Final Program
Sponsored by the SIAM Activity Group on Computational Science and Engineering (CSE) The SIAM Activity Group on CS&E fosters collaboration and interaction among applied mathematicians, computer scientists, domain scientists and engineers in those areas of research related to the theory, development, and use of computational technologies for the solution of important problems in science and engineering. The activity group promotes computational science and engineering as an academic discipline and promotes simulation as a mode of scientific discovery on the same level as theory and experiment. The activity group organizes this conference and maintains a wiki, a membership directory, and an electronic mailing list.
SIAM 2015 Events Mobile App Scan the QR code with any QR reader and download the TripBuilder EventMobile™ app to your iPhone, iPad, iTouch or Android mobile device. You can also visit www.tripbuilder. com/siam2015events
Society for Industrial and Applied Mathematics 3600 Market Street, 6th Floor Philadelphia, PA 19104-2688 USA Telephone: +1-215-382-9800 Fax: +1-215-386-7999 Conference E-mail:
[email protected] Conference Web: www.siam.org/meetings/ Membership and Customer Service: (800) 447-7426 (US & Canada) or +1-215-382-9800 (worldwide) www.siam.org/meetings/cse15
2
2015 SIAM Conference on Computational Science and Engineering
Table of Contents
Mary Ann Leung
General Information............................... 2 Celebrating 15 Years of SIAM CSE...... 5 Prize Award Ceremony.......................... 5 Conference Sponsors............................. 3 Invited Plenary Presentations................. 8 Symposium on Materials .................... 10 Minitutorials......................................... 11 Featured Minisymposia........................ 12 Panels................................................... 14 Career Fair........................................... 16 Professional Development Evening..... 17 Student Days........................................ 19 Workshop Celebrating Diversity.......... 21 AWM Workshop.................................. 21 Mentoring Program.............................. 23 Program Schedule................................ 25 Sunday Poster Session......................... 80 AWM Poster Session........................... 85 Monday Poster Session...................... 111 Speaker and Organizer Index............. 181 Conference Budget.... Inside Back Cover Salt Palace Meeting Room Floor Plan..... .......................................... Back Cover
Anders Logg
Organizing Committee Co-Chairs Hans De Sterck University of Waterloo, Canada
Chris Johnson University of Utah, USA
Lois Curfman McInnes Argonne National Laboratory, USA
Organizing Committee
Sustainable Horizons, USA Chalmers University of Technology, Sweden
Susan Minkoff University of Texas, Dallas, USA
Patrick O’Leary Kitware, USA
Luke Olson University of Illinois at Urbana-Champaign, USA
Steve Parker NVIDIA, USA
Linda Petzold University of California, Santa Barbara, USA
Han-Wei Shen Ohio State University, USA
Feng Xiao Tokyo Institute of Technology, Japan
Irad Yavneh Technion, Israel
SIAM Registration Desk The SIAM registration desk is located in the East Foyer of the The Calvin L. Rampton Salt Palace Convention Center. Registration hours are: Friday, 1:00 PM - 5:00 PM Saturday, 7:15 AM - 5:00 PM Sunday, 7:30 AM - 5:00 PM Monday, 7:30 AM - 5:00 PM Tuesday, 7:45 AM - 5:00 PM Wednesday, 7:45 AM - 2:30 PM
Convention Center Address The Calvin L. Rampton Salt Palace Convention Center
Evrim Acar
100 SW Temple Salt Lake City, UT 84101 USA
University of Copenhagen, Denmark
Phone: +1(385) 468-2222
Tom Bartol Salk Institute, USA
Pavel Bochev Sandia National Laboratories, USA
Hank Childs Lawrence Berkeley National Laboratory and University of Oregon, USA
Vincent Heuveline Heidelberg University, Germany
Child Care To find local child care options in Salt Lake City, please use https://careaboutchildcare.utah.gov/parent/ search.aspx, using the area code 84101, or https://www.guardianangelbaby.com/ home-frame.htm to find reputable child care services.
Corporate Members and Affiliates SIAM corporate members provide their employees with knowledge about, access to, and contacts in the applied mathematics and computational sciences community through their membership benefits. Corporate membership is more than just a bundle of tangible products and services; it is an expression of support for SIAM and its programs. SIAM is pleased to acknowledge its corporate members and sponsors. In recognition of their support, non-member attendees who are employed by the following organizations are entitled to the SIAM member registration rate. Corporate Institutional Members The Aerospace Corporation Air Force Office of Scientific Research Aramco Services Company AT&T Laboratories - Research Bechtel Marine Propulsion Laboratory The Boeing Company CEA/DAM Department of National Defence (DND/ CSEC) DSTO- Defence Science and Technology Organisation ExxonMobil Upstream Research Hewlett-Packard IBM Corporation IDA Center for Communications Research, La Jolla IDA Center for Communications Research, Princeton Institute for Computational and Experimental Research in Mathematics (ICERM) Institute for Defense Analyses, Center for Computing Sciences Lawrence Berkeley National Laboratory Lockheed Martin Los Alamos National Laboratory Mathematical Sciences Research Institute Max-Planck-Institute for Dynamics of Complex Technical Systems Mentor Graphics National Institute of Standards and Technology (NIST) National Security Agency (DIRNSA) Oak Ridge National Laboratory, managed by UT-Battelle for the Department of Energy Sandia National Laboratories Schlumberger-Doll Research Tech X Corporation U.S. Army Corps of Engineers, Engineer Research and Development Center United States Department of Energy *List current January 2015
2015 SIAM Conference on Computational Science and Engineering
Funding Agencies SIAM and the conference organizing committee wish to extend their thanks and appreciation to the U.S. National Science Foundation and the Department of Energy for their support of this conference.
Special Themes: • •
CSE software Big data analytics
Additional Themes: • • • • • • • • • • • • • •
Physics-compatible numerical methods High-accuracy numerical methods Compressed sensing and sparse representation Multiphysics, multiscale, and multilevel methods Reduced-order modeling Visual data analysis Multi-modal methods and data fusion Biomedical computing Computational neuroscience Verification, validation, and uncertainty quantification Extreme-scale and hardware-aware algorithms Modeling and computing complex flows Computational statistics CSE education
Of special note at CSE15: •
•
Increased emphasis on poster sessions, including demos, thematic groups of posters, and new poster prizes 15th Anniversary celebration of SIAMCSE conferences
Leading the applied mathematics community . . . Join SIAM and save! SIAM members save up to $130 on full registration for the 2015 SIAM Conference on Computational Science and Engineering! Join your peers in supporting the premier professional society for applied mathematicians and computational scientists. SIAM members receive subscriptions to SIAM Review, SIAM News, and Unwrapped, and enjoy substantial discounts on SIAM books, journal subscriptions, and conference registrations. If you are not a SIAM member and paid the Non-Member or Non-Member Mini Speaker/Organizer rate to attend the conference, you can apply the difference between what you paid and what a member would have paid ($130 for a Non-Member and $65 for a Non-Member Mini Speaker/Organizer) towards a SIAM membership. Contact SIAM Customer Service for details or join at the conference registration desk. If you are a SIAM member, it only costs $10 to join the SIAM Activity Group on the Computational Science and Engineering (SIAG/ CSE). As a SIAG/CSE member, you are eligible for an additional $10 discount on this conference, so if you paid the SIAM member rate to attend the conference, you might be eligible for a free SIAG/CSE membership. Check at the registration desk. Free Student Memberships are available to students who attend an institution that is an Academic Member of SIAM, are members of Student Chapters of SIAM, or are nominated by a Regular Member of SIAM. Join onsite at the registration desk, go to www. siam.org/joinsiam to join online or download an application form, or contact SIAM Customer Service Telephone: +1-215-382-9800 (worldwide); or 800-447-7426 (U.S. and Canada only) Fax: +1-215-386-7999 E-mail:
[email protected] Postal mail: Society for Industrial and Applied Mathematics, 3600 Market Street, 6th floor, Philadelphia, PA 19104-2688 USA
3
Standard Audio/Visual Set-Up in Meeting Rooms SIAM does not provide computers for any speaker. When giving an electronic presentation, speakers must provide their own computers. SIAM is not responsible for the safety and security of speakers’ computers. The Plenary Session Room will have two (2) screens, one (1) data projector and one (1) overhead projector. Cables or adaptors for Apple computers are not supplied, as they vary for each model. Please bring your own cable/adaptor if using a Mac computer. All other concurrent/breakout rooms will have one (1) screen and one (1) data projector. Cables or adaptors for Apple computers are not supplied, as they vary for each model. Please bring your own cable/adaptor if using a Mac computer. Overhead projectors will be provided for any presenter that ordered one in advance as requested in the acceptance notification. If you have questions regarding availability of equipment in the meeting room of your presentation, please see a SIAM staff member at the registration desk.
E-mail Access A limited number of e-mail stations are available in Room 255 during registration hours. Wireless email access is available throughout the Salt Palace for all conference attendees.
Registration Fee Includes • Admission to all technical sessions • Business Meeting (open to SIAG/CSE members) • Coffee breaks daily • Poster Session and Dessert Reception • Room set-ups and audio/visual equipment • Welcome Reception (held at the Hilton Hotel) In addition, the following events are available to attendees at no additional cost. The events are subsidized by SIAM and are not covered by the registration fees. • • •
Career Fair Champagne Toast at the Welcome Reception Professional Development Evening
4
Job Postings Please check with the SIAM registration desk regarding the availability of job postings or visit http://jobs.siam.org.
2015 SIAM Conference on Computational Science and Engineering
Table Top Displays
Social Media
EDP Sciences
SIAM is promoting the use of social media, such as Facebook and Twitter, in order to enhance scientific discussion at its meetings and enable attendees to connect with each other prior to, during and after conferences. If you are tweeting about a conference, please use the designated hashtag to enable other attendees to follow the conversation and to allow better archiving of our conference discussions. The hashtag for this conference is #SIAMCSE15.
Intel IOS Press
Important Notice to Poster Presenters Poster presenters may set-up their posters between 3:00-5:00 PM on Saturday, March 14. Posters should be set up by 8:00 AM on Sunday, March 15, 2015. Presenters are expected to stand by their posters during their assigned session. Posters should remain on display through 6:30 PM on Monday, March 16, 2015 (two full days). Posters must be removed between 6:30 PM on Monday, and 12:30 PM on Tuesday. Unclaimed posters will be discarded.
SIAM Books and Journals Display copies of books and complimentary copies of journals are available on site. SIAM books are available at a discounted price during the conference. The books table will close at 1:00 PM on Wednesday, March 1.
PIC Math Princeton University Press SIAM Springer The Berkeley MFE
Name Badges A space for emergency contact information is provided on the back of your name badge. Help us help you in the event of an emergency!
SIAM 2015 Events Mobile App Scan the QR code with any QR reader and download the TripBuilder EventMobile™ app to your iPhone, iPad, iTouch or Android mobile device. You can also visit www.tripbuilder.com/siam2015events
Comments? Comments about SIAM meetings are encouraged! Please send to: Cynthia Phillips, SIAM Vice President for Programs (vpp@ siam.org).
Get-togethers “Communication Doctors”: Get your Research Communication Ready for the Press Have a significant research project that you absolutely love, but begin to see people’s eyes glaze over every time you start to talk about it? Visit the “Communication Doctors” at CSE15! During two sessions (held concurrently with the poster sessions) the “Communication Doctors”— representing mathematicians, science communicators and educators—will help make your research story media ready and public friendly. Do you have a scientific question, project or visual that you want to turn into a good story? Bring an elevator pitch, 1-3 minutes long, or a summary or visual of your research work, and the “doctors” will use a variety of skills to break it down for the lay public and media. You will also have a chance to meet and interact with science writer Flora Lichtman, whose work has appeared in the New York Times, NPR’s Science Friday, and Popular Science, among other outlets and publications. Visit the “Communication Doctors” in Room 255 during the poster sessions between 4:30 and 6:30 PM on Sunday, March 15 or Monday, March 16.
•
• •
Welcome Reception Friday, March 13, 6:00-8:00 PM Hilton Salt Lake City Center 255 South West Temple Salt Lake City, UT 84101 Poster Session & Reception Sunday, March 15, 4:30-6:30 PM Poster Session & Reception Monday, March 16, 4:30-6:30 PM
Please Note SIAM is not responsible for the safety and security of attendees’ computers. Do not leave your personal electronic devices unattended. Please remember to turn off your cell phones and other devices during sessions.
Recording of Presentations Audio and video recording of presentations at SIAM meetings is prohibited without the written permission of the presenter and SIAM.
The SIAM 2015 Events Mobile App Powered by TripBuilder® To enhance your conference experience, we’re providing a state-of-the-art mobile app to give you important conference information right at your fingertips. With this TripBuilder EventMobile™ app, you can: • Create your own custom schedule • View Sessions, Speakers, Exhibitors and more • Take notes and export them to your email • View Award-Winning TripBuilder Recommendations for the meeting location • Get instant Alerts about important conference info
2015 SIAM Conference on Computational Science and Engineering
Celebrating 15 Years of SIAM CSE On Saturday, March 14 at 8:30 AM, join Linda R. Petzold as she examines the growth of CSE in SIAM and looks toward some of the challenges and opportunities for the future.
Prize Award Ceremony On Tuesday, March 17, 9:00-9:30 AM the prize award ceremony will take place. The following prizes will be acknowledged and announced: SIAM/ACM Prize in Computational Science and Engineering, CSE15 Poster Prizes and the Computational Engineering (BGCE) Student Paper Prizes.
Poster Blitzes and Plenary Poster Sessions including Minisymposteria New this year are two plenary poster sessions featuring Minisymposteria. Minisymposteria are collections of three or more posters by different presenters grouped around a central theme. Poster sessions will take place in Room 255 on Sunday, March 15, 4:30-6:30 PM and Monday, March 16, 4:30-6:30 PM. Both sessions include Minisymposteria and general posters grouped by category. Poster Blitzes will precede each poster session at 3:40 PM in Room 355. To ensure conference participants have ample time to view posters, presenters are requested to keep their posters on display for two full days, from 8:00 AM Sunday, March 15 through 6:30 PM Monday, March 16. Presenters are expected to stand by their posters during their assigned session. Food and beverage will be provided and posters will be reviewed for poster prizes! These sessions provide a great opportunity to network and are a great prelude to dinner in downtown Salt Lake City!
5
6
2015 SIAM Conference on Computational Science and Engineering
2015 SIAM Conference on Computational Science and Engineering
7
Welcome Reception
To Our Conference Sponsors, Thank You!
Student Careers Panel
Gold sponsor
Gold sponsor
Academic sponsor
General sponsor
8
2015 SIAM Conference on Computational Science and Engineering
Invited Plenary Speakers ** All Invited Plenary Presentations will take place in the Salt Palace Convention Center – Room 355, Level 3**
Saturday, March 14 9:00 AM - 9:45 AM IP1 Graph Data Analytics at Scale: Opportunities and Challenges Nagiza Samatova, North Carolina State University and Oak Ridge National Laboratory, USA 1:30 PM - 2:15 PM IP2 Model Reduction - Trouble with Scales? Wolfgang Dahmen, RWTH Aachen, Germany
Sunday, March 15 8:15 AM - 9:00 AM IP3 Petascale Finite Element Simulation of Real World’s Complex Structure with Billions DOFs Model Shinobu Yoshimura, University of Tokyo, Japan 11:20 AM - 12:05 PM IP4 Extreme-scale Multigrid in Space and Time Robert Falgout, Lawrence Livermore National Laboratory, USA
2015 SIAM Conference on Computational Science and Engineering
Invited Plenary Speakers ** All Invited Plenary Presentations will take place in the Salt Palace Convention Center – Room 355, Level 3**
Monday, March 16 8:15 AM - 9:00 AM IP5 Statistical and Computational Challenges of Constraining Greenhouse Gas Budgets Anna Michalak, Carnegie Institution for Science and Stanford University, USA 11:20 AM - 12:05 PM IP6 Scaling Open Systems for Future Computational Challenges Will Schroeder, Kitware, Inc., USA
Tuesday, March 17 8:15 AM - 9:00 AM IP7 A Calculus for the Optimal Quantification of Uncertainties Houman Owhadi, California Institute of Technology, USA 11:50 AM - 12:35 PM IP8 The Power of Matrix and Tensor Decompositions in Smart Patient Monitoring Sabine Van Huffel, Katholieke Universiteit Leuven, Belgium
Wednesday, March 18 8:15 AM - 9:00 AM IP9 Implications of Numerical and Data Intensive Technology Trends on Scientific Visualization and Analysis James Ahrens, Los Alamos National Laboratory, USA
9
10
2015 SIAM Conference on Computational Science and Engineering
Symposium
Saturday, March 14 NSF-SIAM Symposium on Mathematical and Computational Aspects of Materials Science Part I (MS14): 10:15-11:55 AM Part II (MS40): 2:25-4:05 PM Part III (MS65): 4:35-6:15 PM Room: 251 E The materials science community has embarked on various large-scale computational projects. We mention the materials genome project---a broad survey of thousands of materials---and several deep-mining projects of a few selected materials. These projects offer interesting opportunities for mathematical and computational scientists to advance the state of the art in materials research by developing mathematical models, computational algorithms, and tools for data analysis. The Division of Materials Research (DMR) and Division of Mathematical Sciences (DMS) of the National Science Foundation are encouraging collaborations of mathematicians and materials scientists through several mechanisms (MSII, DMREF, etc.). The speakers in this symposium will highlight opportunities for joint research projects. Organizer: Hans G. Kaper, Argonne National Laboratory and Georgetown University, USA Speakers: MS14 Kaushik Bhattacharya, California Institute of Technology, USA Carme Calderer, University of Minnesota, USA George Crabtree, Argonne National Laboratory, USA Marta Lewicka, University of Pittsburgh, USA MS40 Robert V. Kohn, Courant Institute of Mathematical Sciences, New York University, USA Monica Olvera De La Cruz, Northwestern University, USA Michael Falk, Johns Hopkins University, USA Long-qing Chen, Pennsylvania State University, USA MS65 Richard James, University of Minnesota, USA Sadasivan Shankar, Harvard University, USA Michael S. Vogelius, Rutgers University, New Brunswick, USA Mary Galvin-Donoghue, National Science Foundation, USA
2015 SIAM Conference on Computational Science and Engineering
11
Minitutorials
Sunday, March 15 Minitutorial: Python Visual Analytics for Big Data Part I: 9:10 AM - 10:50 AM Part II: 1:30 PM - 3:10 PM Room: 355 Python is a powerful development, computational, and programming environment due to the wide variety of libraries developed for it, and importantly, the enthusiastic, active development and user community. One of the areas where Python excels is visualization and analysis of big data, due to several high-quality modules for both simple and advanced visual analytics. This tutorial will cover the following big-data visualization capabilities in Python: interactive plotting with IPython, matplotlib, and databases, building web visualizations with Bokeh, and Python integration with VTK and ParaView. Additional information will be provided on mapreduce and NoSQL capabilities. This tutorial is intended for intermediatelevel participants who have a basic understanding of the Python language and development environment (i.e., the student ought to have an understanding of native (and ideally numpy) data structures, file I/O, and is able to develop and run simple programs). Beginner participants are welcome, but Python fundamentals, such as language constructs, “hello world,” and program execution will not be covered in this tutorial. Organizer: Jonathan Woodring, Los Alamos National Laboratory, USA Speakers: Joseph Cottam, Indiana University, USA Jonathan Woodring, Los Alamos National Laboratory, USA Andy R. Terrel, Continuum Analytics, USA Patrick O’Leary, Kitware, Inc., USA
Tuesday, March 17 Minitutorial: Lab Skills for Scientific Computing Part I: 2:15 PM - 3:55 PM Part II: 4:25 PM - 6:05 PM Room: 355 The Software Carpentry project (http://software-carpentry.org) has been teaching basic computing skills to scientists and engineers since 1998. This minitutorial will introduce the tools and techniques that have proven most useful, and show how integrating them can help researchers get more done in less time, and with less pain. This two-part minitutorial will introduce several widely-used practices in software development, explore the empirical evidence showing their benefits (or in some cases the lack thereof), and describe how researchers and research teams can adopt them. Some of the work will be hands-on, so participants are strongly urged to bring a laptop. Warning: real-world examples may be used. Organizer and speaker: Greg Wilson, Mozilla Foundation, USA
12
2015 SIAM Conference on Computational Science and Engineering
Featured Minisymposia Saturday, March 14 MS27 Featured Minisymposium: Fast Multipole Methods Maturing at 30 Years 2:25 PM - 4:05 PM Room: 355 In the last five years or so, research on fast multipole methods has been buzzing and it seems that finally this “top-10” algorithm may be reaching its potential. The mathematicians have always continued making progress and improvements, but lately we see many developments in the computer science aspects of implementing the algorithm with high performance, and the applications arena. This featured minisymposium will present leaders in the field discussing the most resent advances and giving perspectives for the future. It will complement several regular minisymposia being presented at the conference. Organizer: Speakers:
Lorena A. Barba, George Washington University, USA Lorena A. Barba, George Washington University, USA George Biros, University of Texas at Austin, USA Richard Vuduc, Georgia Institute of Technology, USA Lexing Ying, Stanford University, USA
MS52 Featured Minisymposium: Modeling and Computing Complex Flows 4:35 PM - 6:15 PM Room: 355 Computation of unsteady multifluid flows is one of the more challenging problems in computational science. Nevertheless, considerable progress has been made in the development of numerical methods to follow the convoluted motion of complex interfaces separating different fluid phases, for a broad range of governing parameters. These successes are making it possible to pursue even more advanced problems, including additional physics and even large range of scales, as well as making it urgent to develop sophisticated models that take advance of the new data. In this minisymposium we will explore the state of the art and discuss future challenges. Organizer: Speakers:
Gretar Tryggvason, University of Notre Dame, USA Gretar Tryggvason, University of Notre Dame, USA Marianne M. Francois, Los Alamos National Laboratory, USA Marcus Herrmann, Arizona State University, USA Olivier Desjardins, Cornell University, USA
Sunday, March 15 MS104 Featured Minisymposium: Distributed Methods for Optimization 1:30 PM - 3:10 PM Room: 355 A Recent advances in data collection, collaborative filtering, and wireless technology have lead to an unprecedented growth in demand for distributed optimization problems with huge data sets. The resulting optimization problems are characterized by distributed and uncertain information, necessitating computations to be done in a non-traditional environment, with imperfect information, over a communication network, and most importantly without a central entity that has an access to the whole information. This minisymposium focuses on most recent optimization techniques dealing with large data sets and distributed components over possibly uncertain networks. Organizer: Speakers:
Wotao Yin, University of California, Los Angeles, USA Angelia Nedich and Alexander Olshevsky, University of Illinois at Urbana-Champaign, USA Wotao Yin, University of California, Los Angeles, USA Ermin Wei, Massachusetts Institute of Technology, USA Mengdi Wang, Princeton University, USA
2015 SIAM Conference on Computational Science and Engineering
13
Featured Minisymposia Monday, March 16 MS156 Featured Minisymposium: Physics-compatible Numerical Methods 1:30 PM - 3:10 PM Room: 355 Recent advances in data collection, collaborative filtering, and wireless technology have lead to an unprecedented growth in demand Physics-compatible numerical methods are methods that aim to preserve key mathematical and physical properties of continuum physics models in their finite-dimensional algebraic representations. They include methods which preserve prop- erties such as energy, monotonicity, maximum principles, symmetries, and involutions of the continuum models. Examples are mimetic methods for spatial discretizations, variational and geometric integrators, conservative finite-volume and finite- element methods, etc. Research on physics-compatible numerical methods is rapidly becoming a major research thrust across multiple disciplines within the broader area of computational science and engineering. Our principal goal in arranging this minisymosium is to give samples of this flourishing field. Organizer: Speakers:
Mikhail Shashkov, Los Alamos National Laboratory, USA Mikhail Shashkov, Los Alamos National Laboratory, USA Jim E. Morel, Texas A&M University, USA Allen C. Robinson, Sandia National Laboratories, USA Andrew J. Barlow, Atomic Weapons Establishment, United Kingdom
Tuesday, March 17 MS182 Featured Minisymposium: CSE Software 10:00 AM - 11:40 AM Room: 355 A This featured minisymposium reflects recent progress with tools and techniques for developing, adapting, using, and testing CSE software. The first talk addresses an eco system of computational tools for increased efficiency and reliability of scientific workflows. The next talk presents a software framework in C++, MOOSE, which simplifies building advanced multiphysics models. Another talk explains how domain specific languages for PDEs in Python allow automatic generation of adjoint models in C++ with high performance. Finally, the recent Julia language is presented, which combines the expressiveness and convenience of MATLAB/Python with the speed of Fortran/C. Organizer: Speakers:
Hans Petter Langtangen, Simula Research Laboratory and University of Oslo, Norway Derek R. Gaston, Idaho National Laboratory, USA Patrick E. Farrell, University of Oxford, United Kingdom Stephan Karpinski, Massachusetts Institute of Technology, USA Fernando Perez, University of California, Berkeley, USA
Wednesday, March 18 MS255 Featured Minisymposium: Big Data Analytics 10:55 AM - 12:35 PM Room: 355 Big data analytics has become a new paradigm for decision making and knowledge discovery. This minisymposium highlights four major topics, urban, graph, biomedical, and scientific data analytics in this area. Urban data analytics harvest the vast amount of data collected from smart devices and networks to improve the life of citizens. Graphs are used to model relationships between entities and hence play a central role in big data analytics. Biomedical data analytics use data-driven approaches to enable discovery and find new medical cures. Scientific data analytics assists scientists to extract knowledge from simulations aiming to solve grand challenge problems. Organizer: Speakers:
Han-Wei Shen, The Ohio State University, USA Claudio T. Silva, New York University, USA Miriah Meyer, University of Utah, USA Peterka Tom, Argonne National Laboratory, USA Han-Wei Shen, The Ohio State University, USA
14
2015 SIAM Conference on Computational Science and Engineering
Panels Saturday, March 14 12:15 PM - 1:30 PM Student Careers Panel Sponsored by KAUST Room: 355 D The panelists will present an overview of careers in CSE in Academia, Industry and National Laboratories. The session will begin with brief presentations by the panelists, followed by an open discussion and question period with students in the audience. Lunch will be provided. Attendance is limited to current undergraduate and graduate students. Advance sign up is requested during the registration process. If you did not register, but wish to attend, please see a SIAM staff member at the Registration Desk. Organizers: Evrim Acar, University of Copenhagen, Denmark William G. Kolata, SIAM, USA Patrick O’Leary, Kitware, Inc., USA Panelists: Tom Grandine, The Boeing Company, USA David E. Keyes, King Abdullah University of Science & Technology (KAUST), Saudi Arabia Cynthia Phillips, Sandia National Laboratories, USA Jeffrey Saltzman, AstraZeneca, USA
Sunday, March 15 Ask The Program Manager: Funding 12:15 PM - 1:30 PM Room:355 D Program managers from government agencies receive many requests for research funding. What are they really seeking? What makes a research proposal stand out? How can you build a research program that is attractive to these agencies? How can you conduct your research to make the biggest impact and increase your chances of future funding? What opportunities are presently available? We address all these questions and more as a part of this panel discussion. Organizers: Hank Childs, Lawrence Berkeley National Laboratory and University of Oregon, USA Vincent Heuveline, Heidelberg University, Germany Panelists: To Be Announced
Monday, March 16 The Future of CSE as a Discipline 12:15 PM - 1:30 PM Room:355 D Over the past two decades Computational Science and Engineering (CSE) has penetrated the academy, with prominent roles in advancing research and providing interdisciplinary education. However, a combination of disruptive developments -- including extreme-scale computing, big data, and a significant diversification of the applications of CSE in science, technology, and society -is redefining the scope and reach of CSE. This forward-looking panel will explore the future of CSE in a broad academic context. What new “grand challenges” may drive progress in CSE? How can CSE shape the future of new application fields such as computational medicine and biology, computational geoscience, and materials science? Are there opportunities to extend CSE to new areas such as social network analysis, cybersecurity and the social sciences, with mathematics-based large-scale computing rapidly becoming of crucial importance in almost all areas of society? Is the CSE paradigm and focus sufficiently unique and significant to warrant separate programs, graduate degrees, academic departments, and funding streams? Organizer:
David E. Keyes, King Abdullah University of Science & Technology (KAUST), Saudi Arabia
Panelists:
Lorena A. Barba, George Washington University, USA Wolfgang Marquardt, RWTH Aachen University, Germany J. Tinsley Oden, University of Texas at Austin, USA Padma Raghavan, Pennsylvania State University, USA Ed Seidel, University of Illinois at Urbana-Champaign, USA
2015 SIAM Conference on Computational Science and Engineering
15
Panels Tuesday, March 17 Data Science: What is It and How to Teach It 12:45 PM - 2:15 PM Room:355 A Recently, the term Big Data has become ubiquitous. People who can wrangle Big Data are called Data Scientists. According to a number of sources, there is a growing need for people trained as Data Scientists. But what is Data Science? Is Data Science its own field or is it an interdisciplinary mix of Computer Science, Mathematics and Statistics, and Domain Knowledge or perhaps it is really what Statisticians have been doing all along? Since data science at scale involves large-scale computation, what is the relation between data science and computational science in research and education? This panel features leaders in Data Science and Computational Science to discuss the current and future status of Data Science, its relationship to Computational Science, opportunities for Data and Computational Scientists and educating future Data Scientists. Organizers: Christopher Johnson, University of Utah, USA Jeff Phillips, University of Utah, USA Panelists: To Be Announced
Thinking of Writing a Book? 12:45 PM - 2:15 PM Room:355 D Ever thought about writing a book? Ever wondered just what that entails? This session brings together successful authors and publishing staff to discuss the process. Topics of interest will include: • Why and when you should consider writing a book • A step-by-step description of the process, from initial idea to published book • How to choose a publisher • The author/publisher relationship – who does what • Pitfalls to avoid • Tips from successful authors Even if you aren’t currently thinking of becoming an author, this session promises to be lively and engaging! Organizer:
Nicholas J. Higham, University of Manchester, United Kingdom
Panelists:
Daniela Calvetti, Case Western Reserve University, USA Eldad Haber, University of British Columbia, Canada Ralph C. Smith, North Carolina State University, USA Elizabeth Greenspan, SIAM, USA David Marshall, SIAM, USA
16
2015 SIAM Conference on Computational Science and Engineering
Career Fair: Alternatives to Academia Saturday, March 14 Session 1 Session 2 Session 3 Room: 255
10:15-11:55 AM 2:25-4:05 PM 4:35-6:05 PM
The career fair will feature representatives from nonacademic employers from industry and government. These representatives will be prepared to discuss with you the opportunities for internships, postdoctoral appointments and full-time jobs at their organizations. The career fair will feature one morning and two afternoon sessions during which you will have the opportunity to speak with the representatives of the participating organizations. In addition to these sessions, there will be Student Careers Panel over lunch and Professional Development sessions and reception in the evening. SIAM is pleased to share the list of organizations that will participate. This list is current at time of printing. The most up to date list of participants can be found at http://www.siam.org/meetings/cse15/career.php. • Argonne National Laboratory • Boeing • Hewlett-Packard Laboratories • Kitware • Lawrence Berkeley National Laboratory • Lawrence Livermore National Laboratory • MathWorks • MIT Lincoln Laboratory • National Institute of Standards & Technology • NSA • NVIDIA • Oak Ridge National Laboratory • Quantlab
2015 SIAM Conference on Computational Science and Engineering
17
Professional Development Evening Saturday, March 14 Interdisciplinary Research: Challenges and Opportunities 6:30 -9:30 PM Room: 355D 6:30 PM - 7:30 PM Getting Started with Interdisciplinary Research Join us for an evening devoted to developing and sustaining a successful interdisciplinary research program. Several professionals from academia, government, and industry will share their perspective on challenges and rewards of an interdisciplinary research, training opportunities for students and early career researchers, how to identify open problems and find common language with colleagues in a different field and funding opportunities among other topics. The target audience for this event includes early career professionals (i.e., less than five years past last degree), postdocs, and students. However, we are also encouraging participation from the senior professional community during the networking session. Organizers: Maria Emelianenko, George Mason University, USA Cammey Cole Manning, Meredith College, USA Sven Leyffer, Argonne National Laboratory, USA Panelists: Richard Braun, University of Delaware, USA Thomas A. Grandine, The Boeing Company, USA C.T. Kelley, North Carolina State University, USA Carol S. Woodward, Lawrence Livermore National Laboratory, USA 7:30 PM - 8:30 PM Networking Reception 8:30 PM - 9:30 PM Interdisciplinary Research: Sustaining a Successful Program Join us for an evening devoted to developing and sustaining a successful interdisciplinary research program. Several professionals from academia, government, and industry will share their perspective on challenges and rewards of an interdisciplinary research, training opportunities for students and early career researchers, how to identify open problems and find common language with colleagues in a different field and funding opportunities among other topics. The target audience for this event includes early career professionals (i.e., less than five years past last degree), postdocs, and students. However, we are also encouraging participation from the senior professional community during the networking session. Organizers:
Maria Emelianenko, George Mason University, USA Cammey Cole Manning, Meredith College, USA Sven Leyffer, Argonne National Laboratory, USA
Panelists:
Fariba Fahroo, Air Force Office of Scientific Research, USA Omar Ghattas, University of Texas at Austin, USA Jeffrey A. Hittinger, Lawrence Livermore National Laboratory, USA Wil Schilders, Eindhoven University of Technology, Netherlands
18
2015 SIAM Conference on Computational Science and Engineering
2015 SIAM Conference on Computational Science and Engineering
19
Student Days Organized by the SIAM Education Committee, Chaired by Rachel Levy, Harvey Mudd College This annual event is being held at CSE15 because there is no 2015 Annual Meeting. Student Days are for students and about students. Organizers have arranged for activities and sessions where students can meet with both peers and professionals in their field, participate in a career fair, attend an information session on hot areas for jobs and research, and network with SIAM Student Chapters from all over the world. Goals Organized by the SIAM Education Committee (Chaired by Rachel Levy, Harvey Mudd College), Student Days are designed to encourage student participation in SIAM, help students learn more about applied mathematics and computational science as both fields of study and as careers, and provide a forum for emerging mathematicians to learn about their field from professionals. Organizers also hope to encourage those in the learning community to establish new student chapters of SIAM and to promote interaction between students and SIAM leadership. Events and Happenings Activities that will enhance the meeting for students include a special orientation prior to Friday’s Welcome Reception, a career panel and full day career fair featuring non-academic employers on Saturday, professional development sessions and reception Saturday evening, presentations by student chapter representatives on Sunday, a student lounge, sessions on Undergraduate Research on Monday, poster blitz and posters sessions on Monday and Tuesday, and a session for students with selected conference invited speakers on Tuesday.
Student Days Schedule and Other Activities of Interest to Students FRIDAY, MARCH 13 5:00 PM - 6:00 PM Student Orientation (Hilton Hotel) 6:00 PM - 8:00 PM Welcome Reception (Hilton Hotel) SATURDAY, MARCH 14 8:15 AM - 8:30 AM Opening remarks 8:30 AM - 9:00 AM Celebrating 15 years of SIAG/CSE 10:15 AM - 11:55 AM Career Fair 12:15 PM - 1:30 PM PD1 Student Careers Panel lunch 2:25 PM - 4:05 PM Career Fair 4:35 PM - 6:15 PM Career Fair 6:30 PM - 7:30 PM Professional Development session: Getting Started with Interdisciplinary Research 7:30 PM - 8:30 PM Professional Development reception 8:30 PM - 9:30 PM Professional Development session: Interdisciplinary Research: Sustaining a Successful Program SUNDAY, MARCH 15 7:00 AM - 8:00 AM 9:30 AM - 4:30 PM 9:10 AM - 10:50 PM 1:30 PM - 3:10 PM 3:10 PM - 4:30 PM 4:30 PM - 6:30 PM
Student Chapter meeting with SIAM Leadership (by invitation only) Visit the student lounge in Room 255 MS94 Student Days: SIAM Student Chapter Presentations - Part I of II MS119 Student Days: SIAM Student Chapter Presentations - Part II of II Poster Blitz Poster Session
MONDAY, MARCH 16 9:10 AM - 10:50 AM MS146 Student Days: Undergraduate Sessions - Part I of II 9:30 AM - 4:30 PM Visit the student lounge in Room 255 1:30 PM - 3:10 PM MS172 Student Days: Undergraduate Sessions - Part II of II 3:10 PM - 4:30 PM Poster Blitz 4:30 PM - 6:30 PM Poster Session TUESDAY, MARCH 17 10:00 AM - 11:40 AM MS198 Student Days: Combining Disciplines, Techniques, Faculty and Students to Tackle Protein Folding 4:25 PM - 6:05 PM MS247 Student Days: An Informal Meeting with invited speakers
20
2015 SIAM Conference on Computational Science and Engineering
2015 SIAM Conference on Computational Science and Engineering
21
Workshop Celebrating Diversity Organizers:
Raegan Higgins, Texas Tech University, USA Susan E. Minkoff, University of Texas at Dallas, USA Stephen A. Wirkus, Massachusetts Institute of Technology, USA
This annual event is being held at CSE15 this year because there is no 2015 Annual Meeting. The Workshop provides a chance for students to listen to technical talks presented by minority graduate students. The workshop is intended to accomplish several goals: To send a clear, explicit message of enthusiastic welcome and support from SIAM to members of underrepresented groups. The workshop is deliberately held as part of a regular SIAM meeting so that the participants can combine the experiences of attending a regular scientific meeting and a special occasion dedicated to them. To bring together a mixture of people from different levels of age and professional experience, ranging from under graduate students to senior scientists. To provide an opportunity for minority graduate students to present their research. To provide an informal, comfortable setting (a lunch) where all the students can meet applied and computational mathematicians with a wide variety of jobs in academia, national laboratories, industry, and government. SIAM and the Conference Organizing Committee wish to extend their thanks and appreciation to the U.S. National Science Foundation for its support of the Workshop Celebrating Diversity.
Workshop Celebrating Diversity Sessions MS86 MS113 MS140 MS166 MS192 MS216
Computational Advances in Energy Research Fluid Transport Dynamics in Biology and Medicine Modern Computational Modeling in Fluids Computational Science for Current Multidisciplinary Research Problems The System Dynamics of Social and Health Processes using Quantitative Data Sciences Methods Water Resources Management: How to Add it All Up
Association for Women in Mathematics (AWM) Workshop The Workshop for Women Graduate Students and Recent P.h.D.’s. Please visit http://www.awm-math.org/workshops.html for more information.
AWM Sessions MS13 MS39 MS89 MS114 PP8
Career Development: Celebrating Firsts - Lessons from Trailblazers, First Ones, and Only Ones Career Development: Celebrating Firsts - Panel and Roundtable Research Talks by Recent Ph.D’s: Mathematical Modeling and High-Performance Computing for Multiscale and Multiphysics Problems - Part I of II Research Talks by Recent Ph.D.s: Mathematical Modeling and High-performance Computing for Multi-physics and Multi-scale Problems. Part II of II AWM Poster Session (Sunday)
22
2015 SIAM Conference on Computational Science and Engineering
2015 SIAM Conference on Computational Science and Engineering
23
Mentoring Program SIAM CSE15 - Broader Engagement Pilot Mentoring Program The Society of Industrial and Applied Mathematics (SIAM) and Sustainable Horizons Institute are pleased to announce a pilot mentoring program at the 2015 SIAM Conference on Computational Science and Engineering (CSE15). CSE15 is planning a variety of activities to celebrate and support a diverse community. These activities include the Workshop Celebrating Diversity, Professional Development Evening, Association of Women in Mathematics activities, Student Days, Student Careers Panel, and Job Fair. In addition, there will be a mentoring program for a small cohort of CSE15 students, faculty, and professionals.
Mentor Protégé Program at CSE15 There is irrefutable and growing evidence that mentoring promotes academic and career success, especially among those traditionally underrepresented in CSE. Our vision for the Mentor Protégé program is to influence career and professional development long after the conference concludes. The CSE15 Mentor Protégé program includes the following components: • A limited number of Protégés will be recruited from the Workshop Celebrating Diversity and other conference attendees. • Mentors will be recruited from CSE15 attendees and others known to be active in the SIAM and Broader Engagement (BE) communities. • A brief welcome and orientation for Protégés will occur on Friday, March 13, 2015 (Hilton Hotel, Room: Canyon A/B). • An initial mentor-protégé introduction/mixer will occur Saturday March 14, 2015 (Salt Palace Convention Center, Room 255). • A mentoring activity will occur during the Workshop Celebrating Diversity luncheon on Tuesday, March 17, 2015.
Benefits • •
Protégés will benefit from the opportunity to connect with seasoned professionals, get advice, and explore their questions about academic, career, and professional pathways. Mentors will benefit from the opportunity to give back to the community, influence others, and discover and share their own techniques and aspirations.
Mentor Protégé Matching Process and Responsibilities Mentors and Protégés will be matched utilizing demographic and other information about scientific, professional, and personal interests. MentorProtégé pair introductions will occur electronically prior to the conference and in-person during the first event on Saturday. Mentors and Protégé responsibilities include participation in the introduction event on Saturday and the luncheon activity on Tuesday. Optional additional activities will be suggested, such as mentors’ agenda review and session recommendations for their protégés to consider, joint attendance, one-on-one meetings, and continued contact after the conference. Mentor Protégé Tuesday Lunch Activity The Workshop Celebrating Diversity luncheon will provide an excellent opportunity to formally discuss mentoring, suggest activities, and help to establish meaningful mentoring relationships. The luncheon will include a mentoring workshop based on the “Pathways to Success” Workshop: exploring challenges and identifying resources to help people create a vision of success for themselves. If you have any questions about the Mentoring Program, contact: Mary Ann Leung,
[email protected].
24
2015 SIAM Conference on Computational Science and Engineering
2015 SIAM Conference on Computational Science and Engineering
Program Schedule
25
26
2015 SIAM Conference on Computational Science and Engineering
Friday, March 13 Registration 1:00 PM-5:00 PM Room:East Foyer
Saturday, March 14 Registration 7:15 AM-5:00 PM Room:East Foyer
Mentoring Session 4:00 PM-5:00 PM Room:Hilton Hotel - Canyon AB
Student Orientation 5:00 PM-6:00 PM Room:Hilton Hotel - Topaz
Welcome Reception 6:00 PM-8:00 PM Room:Hilton Hotel - Grand Ballroom
Opening Remarks 8:15 AM-8:30 AM Room:355
Saturday, March 14
SP1 Celebrating 15 Years of SIAM CSE 8:30 AM-9:00 AM Room:355 Chair: Christopher Johnson, University of Utah, USA There can be no doubt that SIAM CSE has been a big success! We examine the growth of CSE in SIAM, and more broadly as a discipline, and look toward some of the challenges and opportunities for the future. Linda R. Petzold, University of California, Santa Barbara, USA
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
IP1 Graph Data Analytics at Scale: Opportunities and Challenges 9:00 AM-9:45 AM Room:355 Chair: Hans De Sterck, University of Waterloo, Canada The four V’s of Big Data necessitate fundamentally different data analytics. A promising strategy toward understanding of a complex system’s dynamics and function aims to extract features and relationships between them and to analyze how their evolution causes different functional system responses. Discovery and forecasting of patterns in such feature graphs can provide insights about the vulnerability of our nation’s energy infrastructure to disturbances, the spread of a cyber-security attack, or the anomalies in internode communication in high performance systems. This talk will present some opportunities and challenges in using this strategy for computational science and engineering applications.
Nagiza Samatova North Carolina State University and Oak Ridge National Laboratory, USA
Coffee Break 9:45 AM-10:15 AM Room:255
27
Saturday, March 14
Saturday, March 14
MS1
MS2
Network Science 10:15 AM-11:55 AM
Data-methods for Complex Systems - Part I of III 10:15 AM-11:55 AM
Room:355 Network science is the study of the graphs and networks that arise from current investigations in social networks, biology, power systems, information retrieval, and neuroscience. The goal of this minisymposium is to highlight new developments in this field in terms of algorithms, models, and applications. In particular, the talks will discuss understanding communication patterns in the brain via networks, statistical properties of large-scale heterogenous networks, influencing social networks, and local analysis of massive networks.
Organizer: David F. Gleich Purdue University, USA
Organizer: Tamara G. Kolda Sandia National Laboratories, USA 10:15-10:35 Local Methods in Network Science David F. Gleich, Purdue University, USA 10:40-11:00 Identifying the Largest Entries in Matrix Multiplication Grey Ballard, Tamara G. Kolda, and Ali Pinar, Sandia National Laboratories, USA; C. Seshadhri, University of California, Santa Cruz, USA 11:05-11:25 Network Science of Brain Networks Zoltan Toroczkai, University of Notre Dame, USA 11:30-11:50 Mining Uncertain Networks Evamaria Terzi, Boston University, USA
Room:355 A For Part 2 see MS28 Data-driven methods are transforming the engineering, physical and biological sciences. Indeed, innovative uses of machine learning, dimensionality reduction, sparse sensing, and/or network characterization techniques are allowing for significant advances in engineering designs for the prediction and control of highly complex, often networked, systems. This minisymposium brings together experts who are integrating together one or more of the aforementioned methodologies with the goal of providing transformative analytic tools and algorithms for characterizing the underlying, low-dimensional, dynamics of the complex system.
Organizer: J. Nathan Kutz University of Washington, USA
Organizer: Steven Brunton University of Washington, USA
Organizer: Joshua Proctor Institute for Disease Modeling, USA 10:15-10:35 Self-Tuning Complex Systems J. Nathan Kutz, University of Washington, USA 10:40-11:00 Cluster-based Reducedorder Modelling: From Shear Flows to Engine Tumble Motion Eurika Kaiser, Bernd R. Noack, and Laurent Cordier, CNRS, France; Andreas Spohn, ENS, France; Marc Segond and Markus W Abel, Ambrosys GmbH, Germany; Guillaume Daviller, CERFACS, France; Jan Östh and Sinisa Krajnovic, Chalmers University of Technology, Sweden; Yujun Cao and Jacques Boree, ENSMA, France; Robert K. Niven, University of New South Wales, Australia; Louis N. Cattafesta, Florida State University, USA
continued on next page
28
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS2
MS3
Data-methods for Complex Systems - Part I of III 10:15 AM-11:55 AM
Geometric Methods for Graph Partitioning Part I of II 10:15 AM-11:55 AM
continued
Room:355 D 11:05-11:25 The Impact of L1 optimization in Nonlinear PDE Stanley J. Osher, University of California, Los Angeles, USA 11:30-11:50 Title Not Available at Time of Publication Richard G. Baraniuk, Rice University, USA
For Part 2 see MS29 The graph partitioning problem of finding meaningful clusters in a dataset has a variety of important applications in areas such as machine learning, image analysis, and topic modeling. A variety of new methods have recently been introduced based on ideas from geometric analysis, compressive sensing, spectral graph theory, and nonlinear partial differential equations. This minisymposium aims to bring together mathematicians and scientists working on these methods to share new results and exchange ideas.
Organizer: Braxton Osting University of Utah, USA
Organizer: Dominique Zosso University of California, Los Angeles, USA 10:15-10:35 Modulus of Families of Walks on Graphs Nathan Albin, Kansas State University, USA 10:40-11:00 Graph Directed Topic Modeling Arjuna Flenner, Naval Air Weapons Station, USA; Cristina Garcia-Cardona, Los Alamos National Laboratory, USA 11:05-11:25 Building Graphs to Analyze Big Data Blake Hunter, Claremont McKenna College, USA 11:30-11:50 An Incremental Reseeding Strategy for Clustering Thomas Laurent, Loyola Marymount University, USA
Saturday, March 14
MS4 Adaptive Model Order Reduction - Part I of III 10:15 AM-11:55 AM Room:250 A For Part 2 see MS30 Classical model reduction follows a static approach where the reduced-order model is built once in the offline phase and then remains unchanged in the online phase. Adaptive model reduction breaks with this rigid splitting and adapts the reduced order model online. Such adaptive techniques include localization approaches, where one of several pre-computed local models is selected depending on the current state of the system; dictionary approaches, where the reduced basis is assembled on demand from pre-computed basis vectors; and updating methods, which incorporate new data or other information in the reduced-order model.
Organizer: Benjamin Peherstorfer Massachusetts Institute of Technology, USA
Organizer: David Amsallem Stanford University, USA 10:15-10:35 Online-Adaptive Reduced Bases for Parametric Problems Bernard Haasdonk, University of Stuttgart, Germany 10:40-11:00 Adaptive $h$-refinement for Reduced-order Models via Basis Splitting Kevin T. Carlberg, Sandia National Laboratories, USA 11:05-11:25 A Nonlinear Trust Region Framework for PDE-Constrained Optimization Using ProgressivelyConstructed Reduced-Order Models Matthew J. Zahr, University of California, Berkeley and Stanford University, USA; Charbel Farhat, Stanford University, USA 11:30-11:50 Local/Global Model Order Reduction Techniques for Fracture Pierre Kerfriden, Cardiff University, United Kingdom
2015 SIAM Conference on Computational Science and Engineering
29
Saturday, March 14
Saturday, March 14
Saturday, March 14
MS5
MS6
MS7
State-of-the-art Iterative Solvers for Inverse Problems Part I of II 10:15 AM-11:55 AM
Computational Techniques for Time Dependent Coupled Multiphysics and Multiscale Problems Part I of III 10:15 AM-11:55 AM
Active Subspaces for Dimension Reduction: Theory and Applications Part I of III 10:15 AM-11:55 AM
Room:250 C
For Part 2 see MS33 Parameter studies like uncertainty quantification and optimization are infeasible for complex simulations with more than a handful of inputs. When a simulation contains several inputs, the engineer may seek an alternative parameterization with fewer variables to enable such studies. Active subspaces are an emerging set of tools for dimension reduction in complex models. The active subspace is a set of directions in the space of input variables that correspond to greater change, on average, in a quantity of interest. This minisymposium will explore (i) recent developments in algorithms that exploit active subspaces and (ii) applications from across engineering disciplines.
Room:250 B For Part 2 see MS31 Efficient linear solvers are a critical component in the large-scale implementations of algorithms for inverse problems. Examples include solvers for the Newton/Gauss-Newton step and solving the discretized PDEs which comprise the “forward problem”. This session will focus on advances in iterative solvers and preconditioners for linear and nonlinear inverse problems. Applications of the algorithms include diverse areas ranging from medical imaging to environmental sciences.
Organizer: Arvind Saibaba Tufts University, USA
Organizer: Misha E. Kilmer Tufts University, USA
Organizer: Eric Miller Tufts University, USA 10:15-10:35 A Bound-Plus-Equality Constrained Quadratic Minimization Algorithm for Inverse Problems Johnathan M. Bardsley, University of Montana, USA; Marylesa Howard, National Security Technologies, LLC, USA 10:40-11:00 Statistically Motivated Preconditioners and Stopping Criteria for Biomedical Inverse Problems Daniela Calvetti and Erkki Somersalo, Case Western Reserve University, USA 11:05-11:25 Numerical Implementation of a New Class of Forward-BackwardForward Diffusion Equations for Image Restoration James V. Lambers, University of Southern Mississippi, USA; Patrick Guidotti, University of California, Irvine, USA; Yunho Kim, Ulsan National Institute of Science and Technology, South Korea 11:30-11:50 Recycling Krylov Subspaces for Parametric Linear Systems Arising from Hyperspectral Diffuse Optical Tomography Arvind Saibaba, Misha E. Kilmer, and Eric Miller, Tufts University, USA
For Part 2 see MS32 The simulation of multiphysics and multiscale systems is important for a wide range of applications including fluid structure interaction and conjugate heat transfer. Of particular interest is the formulation of solvers built by reusing codes for the various subproblems for which it is common to use ad hoc coupling or time integration schemes. This minisymposium discusses new and practical approaches to advance the state of the art, and to help bridge the gap between researchers working in the various fields. Important aspects include time adaptivity, high order methods, multirate schemes, stability, coupling conditions, and the overall solver performance.
Organizer: Jeffrey W. Banks Rensselaer Polytechnic Institute, USA
Room:250 D
Organizer: Paul Constantine Colorado School of Mines, USA
10:15-10:35 Multirate GARK Schemes Michael Guenther, Bergische Universität, Germany; Adrian Sandu, Virginia Tech, USA
10:15-10:35 Active Subspaces in Theory and Practice Eric Dow, Massachusetts Institute of Technology, USA
10:40-11:00 The Dirichlet-Neumann Iteration and Unsteady Thermal Fluid Structure Interaction Philipp Birken and Azahar Monge Sanchez, Lund University, Sweden
10:40-11:00 Envelopes: Subspace Methods for Efficient Estimation in Multivariate Statistics Dennis Cook, University of Minnesota, USA
11:05-11:25 Partitioned Fluid-Structure Interaction on Massively Parallel Systems Florian Lindner and Miriam Mehl, Universität Stuttgart, Germany; Benjamin Uekermann, Technische Universität München, Germany 11:30-11:50 Multi-level Acceleration of Strongly Coupled Fluid-structure Interaction with Manifold Mapping David Blom, Alexander H. van Zuijlen, and Hester Bijl, Delft University of Technology, Netherlands
11:05-11:25 Order Determination for Dimension Reduction Using An Alternating Pattern of Spectral Variability Bing Li and Wei Luo, Pennsylvania State University, USA 11:30-11:50 Mathematical Foundations of Subspace Selections Massimo Fornasier, Technical University of Munich, Germany; Valeriya Naumova, Simula Research Laboratory, Norway
30
Saturday, March 14
MS8 Advances in Algorithms for Uncertainty Quantification in Large-scale Inverse Problems - Part I of III 10:15 AM-11:55 AM Room:250 E For Part 2 see MS34 Quantification of uncertainty in largescale inverse problems governed by partial differential equations presents significant challenges due to computationally expensive parameter-to-observable maps and highdimensional parameter spaces. This minisymposium presents recent advances in algorithms that make UQ for large-scale linear and nonlinear inverse problems tractable by exploiting problem structure.
Organizer: Georg Stadler Courant Institute of Mathematical Sciences, New York University, USA
Organizer: George Biros University of Texas at Austin, USA
Organizer: Omar Ghattas University of Texas at Austin, USA
Organizer: Youssef M. Marzouk Massachusetts Institute of Technology, USA 10:15-10:35 Conditions for Successful Data Assimilation in High Dimensions Matthias Morzfeld, Lawrence Berkeley National Laboratory, USA; Alexandre Chorin, University of California, Berkeley, USA 10:40-11:00 Filtering Unstable Quadratic Dissipative Systems Kody Law, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Andrew Stuart, Daniel SanzAlonso, and Abhishek Shukla, University of Warwick, United Kingdom 11:05-11:25 Ensemble Methods for Large-Scale PDE-Constrained Bayesian Inverse Problems Kainan Wang and Tan Bui-Thanh, University of Texas at Austin, USA 11:30-11:50 High Dimensional NonGaussian Bayesian Inference with Transport Maps Alessio Spantini and Youssef M. Marzouk, Massachusetts Institute of Technology, USA
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
MS9
Saturday, March 14
MS10
Matrix-Free Methods for Large-Scale Optimization and Applications 10:15 AM-11:55 AM
Embedded Boundary and Interface Techniques Part I of II 10:15 AM-11:55 AM
Room:250 F
Room:251 A
For Part 2 see MS35 Many applications can be formulated as large-scale optimization problems, including inverse problems, medical and seismic imaging, classification in machine learning, data assimilation in weather prediction, and matrix decompositions. For all of them, explicit modeling is prohibitive and matrixfree methods are essential for competitive performance. First-order methods have proven widely successful in recent years. However, recent developments suggest that matrix-free second-order methods, such as interior-point methods, can be competitive. The goal of the minisymposium is to explore those ideas, compare to first-order methods, and promote further research and collaboration between the optimization and applied communities in developing largescale matrix-free methods.
For Part 2 see MS36 Many problems in science and engineering are modeled by partial differential equations (PDEs) in domains with irregular geometry and/or with interfaces. Development of special algorithms are required to restore the accuracy of the numerical schemes near arbitrary boundaries/or interfaces. Recently, much of the progress has been made for designing highly accurate and efficient numerical methods for such problems which employ only simple Cartesian grids. Developments and applications of XFEM, Immersed Interface Methods and Difference Potentials Methods, as well as open questions in the field will be discussed in details.
Organizer: Dominique Orban École Polytechnique de Montréal, Canada
Organizer: Aleksandr Aravkin IBM T.J. Watson Research Center, USA 10:15-10:35 Matrix Free Methods for Large-Scale Nonlinear Constrained Optimization James V. Burke, University of Washington, USA 10:40-11:00 Title Not Available at Time of Publication Michael Friedlander, University of California, Davis, USA 11:05-11:25 Anatomy of a MatrixFree Interior-Point Solver for Convex Optimization Dominique Orban, École Polytechnique de Montréal, Canada 11:30-11:50 Matrix-Free Interior-Point Method for Large Scale Machine Learning Problems Aleksandr Aravkin, IBM T.J. Watson Research Center, USA
Organizer: Yekaterina Epshteyn University of Utah, USA
Organizer: Gunilla Kreiss Uppsala University, Sweden 10:15-10:35 Difference Potentials Method for Parabolic Models in Irregular Domains Jason Albright and Yekaterina Epshteyn, University of Utah, USA 10:40-11:00 Multidimensional Embedded Finite Difference Methods which Satisfies Energy Estimates Adi Ditkowsky, Tel Aviv University, Israel 11:05-11:25 High-Order Accurate Difference Potentials Methods for the Stokes--Darcy Problem Yekaterina Epshteyn and Kyle R. Steffen, University of Utah, USA 11:30-11:50 High Order Cut Finite Elements Methods August Johansson, Simula Research Laboratory, Norway; Mats G. Larson, Umeå University, Sweden
2015 SIAM Conference on Computational Science and Engineering
31
Saturday, March 14
Saturday, March 14
Saturday, March 14
MS11
MS12
MS13
Physics-compatible Discretization on Multiphysics Systems and Efficient Multilevel Solvers - Part I of II 10:15 AM-11:55 AM
Innovative Algorithms for Eigenvalue and Singular Value Decomposition Part I of III 10:15 AM-11:55 AM
Room:251 B
Room:251 C
AWM Workshop - Career Development: Celebrating Firsts - Lessons from Trailblazers, First Ones, and Only Ones 10:15 AM-11:55 AM
For Part 2 see MS37 Design of the appropriate discretization for the PDE systems can oftentimes be done by taking into account the detailed characteristics of the PDE models in relation to the underlying Physics. In many cases, efficient multilevel iterative solvers can be developed based on such discretization. Therefore, such discretization is typically solver-friendly. The minisymposium is to disseminate the stateof-art developments in physics-compatible discretization on multi-physics systems and the corresponding efficient multilevel solvers.
For Part 2 see MS38 The minisymposium will focuses on the first steps taken on the development of novel software methodologies and algorithm for the next generation of HPC systems. Some scale challenges will be addressed; the goal is to close the “application-architecture peak performance gap” by exploring algorithms and runtime improvements that will enable key science applications to better exploit the architectural features of the extreme-scale systems. The contributed talks will cover new approaches that can overcome the limitations of existing dense/sparse eigensolver libraries on platforms that require fine granularity and memoryaware computational tasks combined with asynchronism in parallel execution.
Organizer: Yunrong Zhu Idaho State University, USA
Organizer: Young-Ju Lee Texas State University, USA 10:15-10:35 Robust Multilevel Preconditioners for Elliptic Problems with Discontinuous Coefficients Yunrong Zhu, Idaho State University, USA 10:40-11:00 Finite Element Multigrid Framework for Mimetic Finite Difference Discretizations Francisco José Gaspar, University of Zaragoza, Spain; Xiaozhe Hu, Tufts University, USA; Carmen Rodrigo, University of Zaragoza, Spain; Ludmil Zikatanov, Pennsylvania State University, USA 11:05-11:25 Adaptive Regularization Strategies for Nonlinear PDE Sara Pollock, Texas A&M University, USA 11:30-11:50 New Multigrid Methods for Saddle Point Problems Susanne Brenner, Louisiana State University, USA; Hengguang Li, Wayne State University, USA; Li-yeng Sung, Louisiana State University, USA
Organizer: Azzam Haidar University of Tennessee, Knoxville, USA
Organizer: Piotr Luszczek University of Tennessee, Knoxville, USA
Organizer: Stanimire Tomov University of Tennessee, Knoxville, USA 10:15-10:35 Efficient Eigensolver Algorithm on Accelerator-Based Architecture Azzam Haidar, Piotr Luszczek, and Stanimire Tomov, University of Tennessee, Knoxville, USA 10:40-11:00 Towards Materials Design with Extreme-Scale Quantum Simulations Thomas C. Schulthess, Anton Kozhevnikov, and Solcà Raffaele, ETH Zürich, Switzerland 11:05-11:25 GPGPU Acceleration of the Ams Eigensolver Using Magma Mintae Kim, Luis Crivelli, Michael Wood, Cristian Ianculescu, and Vladimir Belsky, Dassault Systèmes, USA 11:30-11:50 High-Performance Computation of Pseudospectra Jack L. Poulson, Stanford University, USA
Room:251 D For Part 2 see MS39 Sponsored by Association for Women in Mathematics (AWM) The historical awarding of the Fields Medal to a female mathematician reminds us that women in mathematics, engineering, and science continue to break barriers and blaze trails in academia, industry, national laboratories and institutes. What does it take to be the first woman graduate or tenured professor? What does it take to be a trailblazer at the forefront of a research field? What does it take to succeed when you are the only woman in your department? These amazing individuals will share their stories, strategies, and lessons learned as we celebrate the success of past and future firsts.
Organizer: Elebeoba May University of Houston, USA
Organizer: Hoa Nguyen Trinity University, USA
Organizer: Yekaterina Epshteyn University of Utah, USA Speakers To Be Announced
32
Saturday, March 14
MS14 NSF-SIAM Symposium on Mathematical and Computational Aspects of Materials Science Part I of III 10:15 AM-11:55 AM Room:251 E For Part 2 see MS40 The materials science community has embarked on various large-scale computational projects. We mention the materials genome project---a broad survey of thousands of materials---and several deepmining projects of a few selected materials. These projects offer interesting opportunities for mathematical and computational scientists to advance the state of the art in materials research by developing mathematical models, computational algorithms, and tools for data analysis. The Division of Materials Research (DMR) and Division of Mathematical Sciences (DMS) of the National Science Foundation are encouraging collaborations of mathematicians and materials scientists through several mechanisms (MSII, DMREF, etc.). The speakers in this symposium will will highlight opportunities for joint research projects.
Organizer: Hans G. Kaper Argonne National Laboratory and Georgetown University, USA 10:15-10:35 Opportunities and Challenges in First Principles Models of Materials Kaushik Bhattacharya, California Institute of Technology, USA
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
MS15
MS16
Efficient High-order Numerical Methods for Nonlinear PDEs - Part I of II 10:15 AM-11:55 AM
Theory, Computation and Experiments in Biofluids Part I of II 10:15 AM-11:55 AM
Room:251 F
Room:254 A
For Part 2 see MS64 Recent years have seen growing trends in the development of efficient high-order numerical schemes for nonlinear PDEs including hyperbolic equations, HamiltonJacobi equations, and quantum and kinetic models. Many challenges in computational efficiency, accuracy, multi-scales, and theoretical foundations are actively addressed for various applications. This minisymposium will bring together researchers to exchange ideas and recent developments in these fields, with applications such as rarefied gas dynamics, semi-conductor simulation, optimal transportation and optimal control.
For Part 2 see MS42 Over the last two decades the study of systemic biofluids and their interactions with engineering therapeutics and implants has rapidly developed into an interdisciplinary approach of utilizing Mathematics, Bioengineering, Biomedical, and the Biological Sciences to address the complexities and challenges of this field. Eight speakers in this two part minisymposium will present their research spanning mathematical analysis, algorithm development, complex simulations of biological systems and coordination between modeling and experiments. The diversity of these presentations shows the importance of interdisciplinary research on systemic biofluids to understand its wide range of domains from basic processes of fluid dynamics to technological applications.
Organizer: Yingda Cheng Michigan State University, USA
Organizer: Fengyan Li Rensselaer Polytechnic Institute, USA 10:15-10:35 High-order Discontinuous Galerkin Methods for Some Kinetic Models Fengyan Li, Rensselaer Polytechnic Institute, USA 10:40-11:00 High Order SemiLagrangian Discontinuous Galerkin Schemes for First and Second Order PDEs Olivier Bokanowski, Université Paris-Diderot, France
10:40-11:00 New Liquid-Crystal Based Models and Technologies Carme Calderer, University of Minnesota, USA
11:05-11:25 Convergence of SemiDiscrete Stationary Wigner Equation with Inflow Boundary Conditions Tiao Lu, Ruo Li, and Zhangpeng Sun, Peking University, China
11:05-11:25 Opportunities in Computational Science: Genomes, Mesoscale and Closing the Loop George Crabtree, Argonne National Laboratory, USA
11:30-11:50 Self-Organized Hydrodynamics in An Annular Domain: Modal Analysis and Nonlinear Effects Hui Yu, Universite de Toulouse, France
11:30-11:50 Problems in Pattern Formation, Geometry and Design of Materials Marta Lewicka, University of Pittsburgh, USA
Saturday, March 14
Organizer: Hoa Nguyen Trinity University, USA
Organizer: John Carrola Southwest Research Institute, USA 10:15-10:35 Modeling Escherichia Coli Chemotaxis in a Fluid Hoa Nguyen, Trinity University, USA 10:40-11:00 The Role of Intraclot Transport in the Dynamics of Platelet Deposition and Coagulation Under Flow Aaron L. Fogelson, University of Utah, USA; Karin Leiderman, University of California, Merced, USA 11:05-11:25 Modeling Cardiac Electro-Fluid-Mechanical Interaction Boyce Griffith, University of North Carolina at Chapel Hill, USA 11:30-11:50 An Integrative Model of Lamprey Locomotion Using the Immersed Boundary Method Christina Hamlet, Tulane University, USA; Eric Tytell, Tufts University, USA; Lisa J. Fauci, Tulane University, USA
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
33
Saturday, March 14
MS18
MS19
Parallel Programming Models, Algorithms and Frameworks for Extreme Computing - Part I of III 10:15 AM-11:55 AM
Advanced Algorithms in Computational Electromagnetics Part I of II 10:15 AM-11:55 AM
Computational Methods for Kinetic Equations and Related Models - Part I of V 10:15 AM-11:55 AM Room:150 AB
Room:254 B
Room:254 C
For Part 2 see MS43 Multicore/manycore processors and accelerators are universally available as both collections of homogeneous standard microprocessors and as attached heterogeneous co-processors. Application and library software developers may often effectively use these processors and some general approaches have emerged. It is widely recognized that careful design of software and data structures, with effective memory management, are the most critical issues to obtain scalable optimized performance on those systems. In these minisymposia we discuss current experiences and development of applications, libraries and frameworks using a variety of hardware. Speakers will address performance results and software design.
For Part 2 see MS44 The minisymposium focuses on advanced numerical algorithms and high performance simulations for electromagnetic systems, including photonics crystals, plasmonics, solar cells, and quantum electronic devices. Main topics include high-order PDE/ODE discretizations, Green function, integral equation, boundary element method, and density matrix approaches. Scalable algorithms and efficient preconditioning strategies are also discussed for solving largescale systems on the advanced computing platforms.
For Part 2 see MS45 Kinetic descriptions play an important role in a variety of physical, biological, and even social applications. Unfortunately, the large phase space associated with the kinetic description has in the past made simulations impractical in most settings. However, recent advances in computer resources and numerical algorithms are making kinetic models more tractable, and this trend is expected to continue in the future. The purpose of this minisymposium is to report on the continuing progress on numerical analysis and computational science for kinetic equations. It brings together researchers from applied mathematics, computational science, physics, and engineering communities to discuss their work and exchange ideas.
MS17
Organizer: Kengo Nakajima University of Tokyo, Japan
Organizer: Michael Heroux Sandia National Laboratories, USA
Organizer: Serge G. Petiton CNRS/LIFL and INRIA, France 10:15-10:35 Opportunities and Challenges in Developing and Using Scientific Libraries on Emerging Architectures Michael Heroux, Sandia National Laboratories, USA 10:40-11:00 ppOpen-APPL/HEXA: A Framework for Development of Parallel FEM/FVM Applications on Intel Xeon Phi Kengo Nakajima, University of Tokyo, Japan 11:05-11:25 On First Experiments for Nuclear Engineering Applications on Intel Xeon Phi Christophe Calvin, CEA Saclay, France
Organizer: MiSun Min Argonne National Laboratory, USA
Organizer: David P. Nicholls University of Illinois, Chicago, USA
Organizer: Martin Frank
10:15-10:35 Fast Solvers for Wave Propagation and Scattering by General Structures Oscar P. Bruno, California Institute of Technology, USA
RWTH - Aachen University of Technology, Germany
10:40-11:00 Generalized Combined Sources Integral Equations for Helmholtz Transmission Problems Catalin Turc, New Jersey Institute of Technology, USA 11:05-11:25 Scalable Algorithms for Density Matrix Calculations of Cavity Quantum Electrodynamic Systems Matthew Otten, Cornell University, USA 11:30-11:50 Electromagnetic Power Absorption and Plasmon Resonances on Rough Conducting Surfaces Carlos A. Perez-Arancibia and Oscar P. Bruno, California Institute of Technology, USA
Organizer: Cory Hauck Oak Ridge National Laboratory, USA
Organizer: Ryan G. McClarren Texas A&M University, USA
Organizer: Jingmei Qiu University of Houston, USA
Organizer: Jeffrey A. Willert Los Alamos National Laboratory, USA 10:15-10:35 Generalized Radiative Transfer: Accounting Accurately for Unresolved Variabilities at No Computational Cost, Yet Without Homogenization Anthony B. Davis and Feng Xu, California Institute of Technology, USA
11:30-11:50 Managing Portability for ASC Applications Jeff Keasler and Richard Hornung, Lawrence Livermore National Laboratory, USA continued on next page
34
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
Saturday, March 14
MS19
MS20
MS21
Computational Methods for Kinetic Equations and Related Models - Part I of V 10:15 AM-11:55 AM
Stochastic Simulation of Turbulent Flows Using Onedimensional Turbulence 10:15 AM-11:55 AM
Career Fair: Alternatives to Academia - Part I of III 10:15 AM-11:55 AM
continued
Room:150 DE
For Part 2 see MS47 The career fair will feature representatives from nonacademic employers from industry and government. These representatives will be prepared to discuss with you the opportunities for internships, postdoctoral appointments and full-time jobs at their organizations.
10:40-11:00 Quadrature-Based Moment Methods for Radiation Transport Rodney O. Fox, Iowa State University, USA; Cory Hauck, Oak Ridge National Laboratory, USA; Ming Tse P. Laiu, University of Maryland, College Park, USA; Frederique Laurent, Ecole Centrale de Paris, France; Marc Massot, CNRS, France 11:05-11:25 Stability of PN Approximations for the Radiative Transfer Equation in the Free Streaming Limit Matthias Schlottbom, Institute for Computational and Applied Mathematics, Germany; Herbert Egger, Technische Universität Darmstadt, Germany 11:30-11:50 On Combining Moment Methods and Discrete-VelocitySchemes for Solving the Boltzmann Equation Manuel Torrilhon, RWTH Aachen University, Germany
Simulation of turbulent flows is an important and challenging problem due to the wide range of length and time scales involved. One-dimensional turbulence (ODT) is a stochastic model that resolves a full range of length and time scales while remaining computationally affordable. ODT has been applied to a wide range of multiphysics problems including buoyant flows, combustion, and multiphase flows. This minisymposium presents recent advances in stochastic computational science using ODT. An overview of ODT is presented, with applications to soot formation in flames, multiphase reacting and nonreacting flows, and extension of ODT to three dimensions by coupling with large eddy simulation.
Organizer: David O. Lignell Brigham Young University, USA 10:15-10:35 One-dimensional Turbulence Simulation: Overview and Application to Soot Formation in Nonpremixed Flames David O. Lignell and Victoria Lansinger, Brigham Young University, USA; John C. Hewson, Sandia National Laboratories, USA 10:40-11:00 Particle-Scalar Field Interactions in One-Dimensional Turbulence John C. Hewson, Sandia National Laboratories, USA; Guangyuan Sun and David O. Lignell, Brigham Young University, USA 11:05-11:25 Multiphase Turbulent Reacting Flow Simulations Using ODT James C. Sutherland and Babak Goshayeshi, University of Utah, USA 11:30-11:50 ODTLES: A Multiscale Approach for Highly Turbulent Flows Christoph Glawe, Technical University Cottbus, Germany; Heiko Schmidt, Brandenburgische Technische Universität Cottbus, Germany; Alan Kerstein, Consultant, USA
Room:255
Organizer: William G. Kolata SIAM, USA
Organizer: Kristin O’Neill SIAM, USA This is the most current list at time of printing. • Argonne National Laboratory • Boeing • Hewlett-Packard Laboratories • Kitware • Lawrence Berkeley National Laboratory • Lawrence Livermore National Laboratory • MathWorks • MIT Lincoln Laboratory • National Institute of Standards & Technology • NSA • NVIDIA • Oak Ridge National Laboratory • Quantlab The most current list of participating companies is available at www.siam. org/meetings/cse15/career.php.
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
MS22 Computational Methods for Periodic Flows and their Applications 10:15 AM-11:55 AM Room:260 A
Saturday, March 14
MS23 Recent Developments of DG/WENO Methods for Partial Differential Equations - Part I of II 10:15 AM-11:55 AM
35
Saturday, March 14
MS24 Advances in Radial Basis Function and Other Meshfree Methods Part I of V 10:15 AM-11:55 AM
Room:260 B
Room:151 AB
University of California, Merced, USA
For Part 2 see MS49 This minisymposium is to bring people together to discuss the recent advances and exchange ideas in the algorithm design of discontinuous Galerkin (DG) method and weighted essentially nonoscillatory (WENO) method, including the implementation, numerical analysis of those high-order numerical methods for solving partial differential equations. In the minisymposium, the speakers will apply those high-order numerical methods to computational fluid, biology and physics, etc. This minisymposium is a good opportunity for people to discuss with researchers from different areas, and explore more applications and future research collaborations.
For Part 2 see MS50 Meshfree methods have many inherent properties that make them useful for a variety of problems in science and engineering, from fitting data to numerically solving differential equations. Such methods offer geometric flexibility, non-uniform resolution, and advantageous trade-offs between accuracy and computational costs. This minisymposium focuses primarily on methods based on radial basis functions and other more general kernels. The talks will address recent advances in the application of these methods to large-scale problems in biology, geophysics, image processing, and finance, as well as theoretical advances in the methods themselves.
Organizer: Karin Leiderman
Organizer: Yang Yang
Organizer: Varun Shankar
University of California, Merced, USA
Michigan Technological University, USA
University of Utah, USA
10:15-10:35 Computation of the Regularized Image Systems for DoublyPeriodic Brinkman Flow in the Presence of a Wall Hoang-Ngan Nguyen and Karin Leiderman, University of California, Merced, USA
Organizer: Juan Cheng
Organizer: Grady B. Wright
Institute of Applied Physics and Computational Mathematics, China
Boise State University, USA
Many biological and physical flows at lowReynolds number exhibit periodic behavior. One modeling approach to investigate these flows is the use of periodic Green’s functions, or, the periodic summation of the corresponding Green’s functions in free space. While simple in concept, the practical computation of periodic Green’s functions is nontrivial since brute force summation can lead to divergent or slowly convergent sums. To make numerical computation feasible, other approaches such as Ewald summation methods must also be applied.The aim of this minisymposium is to share insight into computational methods for periodic flows and their applications.
Organizer: Hoang-Ngan Nguyen
10:40-11:00 Accelerated Boundary Integral Simulations for Fluid-Structure Interactions in Periodic Stokes Flow Anna-Karin Tornberg, KTH Royal Institute of Technology, Sweden 11:05-11:25 A Numerical Method for Doubly-Periodic Stokes Flow Near a Wall Franz M. Hoffmann, Tulane University, USA 11:30-11:50 Cilia Beating Patterns are not Hydrodynamically Optimal Hanliang Guo, University of Southern California, USA; Janna C. Nawroth, Harvard University, USA; Yang Ding, Beijing Computational Science Research Center, China; Eva Kanso, University of Southern California, USA
10:15-10:35 Symmetry-Preserving Conservative Lagrangian Scheme for Compressible Euler Equations in TwoDimensional Cylindrical Coordinates Juan Cheng, Institute of Applied Physics and Computational Mathematics, China 10:40-11:00 Superconvergent HDG Methods for Third-Order Equations in One-Space Dimension Bo Dong and Yanlai Chen, University of Massachusetts, Dartmouth, USA; Bernardo Cockburn, University of Minnesota, Minneapolis, USA 11:05-11:25 A Local Discontinuous Galerkin Scheme for the Patlak-KellerSegel Chemotaxis Model Yang Yang, Michigan Technological University, USA; Xingjie Li and Chi-Wang Shu, Brown University, USA 11:30-11:50 Optimal Error Estimates for Discontinuous Galerkin Methods Based on Upwind Biased Fluxes for Linear Hyperbolic Equations Xiong Meng, University of East Anglia, United Kingdom
10:15-10:35 Guidelines to Modeling the Navier-Stokes and Euler Equations with RBF-FD Natasha Flyer, National Center for Atmospheric Research, USA; Louis Wicker, National Oceanic and Atmospheric Administration, USA; Gregory Barnett, University of Colorado Boulder, USA 10:40-11:00 A High-Order RBF-Based Leray Projection Method for the Incompressible Stokes and NavierStokes Equations Edward Fuselier, High Point University, USA; Varun Shankar, University of Utah, USA; Grady B. Wright, Boise State University, USA 11:05-11:25 A Novel Elliptic Solver Based on RBF-Finite Differences for Understanding the Earth’s Electric System Victor Bayona and Natasha Flyer, National Center for Atmospheric Research, USA 11:30-11:50 The Fast Orthogonal Gradients Method: An RBF Method for Solving PDEs on Point Cloud-Defined Surfaces Cecile M. Piret, Université Catholique de Louvain, Belgium
36
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS25
MS26
Applications and Uses for Tensors 10:15 AM-11:55 AM
Challenges in Computational Cardiac Electrophysiology Part I of II 10:15 AM-11:55 AM
Room:151 DE An in-depth examination of uses for tensors in different applications and areas of research.
Organizer: Martin D. Schatz University of Texas at Austin, USA 10:15-10:35 A Framework for Distributed Tensor Computations Martin D. Schatz, University of Texas at Austin, USA; Tamara G. Kolda, Sandia National Laboratories, USA; Robert A. van de Geijn, University of Texas at Austin, USA 10:40-11:00 Exploiting Multiple Tensor Symmetries though Block Diagonalization Charles Van Loan, Cornell University, USA 11:05-11:25 Tensor Computation for Chemistry and Material Science Justus Calvin and Edward F. Valeev, Virginia Tech, USA 11:30-11:50 Distributed Contraction of Tensors P. Sadayappan and Samyam Rajbhandari, The Ohio State University, USA; Sriram Krishnamoorthy, Pacific Northwest National Laboratory, USA; Pai-Wei Lai, The Ohio State University, USA
Room:151 G For Part 2 see MS51 The use of computer simulation to model cardiac electrophysiology, from the cellular level to the whole-organ, has the potential to revolutionise mechanistic understanding, diagnosis and treatment of arrhythmias. These models are often complex and computationally demanding, and integrating the available electrical and imaging data with its inherent uncertainty makes direct clinical utility particularly challenging. This minisymposium highlights the latest advancements and new approaches to modelling cardiac electrophysiology across multiple scales, including imaging and data integration, which will pave the way for developing clinically tractable simulation environments to aid patient treatment in the future.
Organizer: Chris Cantwell Imperial College London, United Kingdom
Organizer: Frank B. Sachse University of Utah, USA
Organizer: Richard Clayton University of Sheffield, United Kingdom 10:15-10:35 Multi-Scale Modeling of the Failing Heart: from Molecule to Patient Andrew D. McCulloch, Christopher Villongco, Adarsh Krishnamurthy, and David Krummen, University of California, San Diego, USA 10:40-11:00 Towards Multidomain Modeling of Cardiac Electrophysiology Frank B. Sachse, Thomas Seidel, and Jan Christoph Edelmann, University of Utah, USA; Gunnar Seemann, Karlsruhe Institute of Technology, Germany
continued in next column
11:05-11:25 The Role of Microdomains and Ephaptic Coupling in Cardiac Action Potential Propagation James P. Keener, University of Utah, USA 11:30-11:50 Strongly Scalable Numerical Approaches for Modeling Cardiac Electromechanics at High Spatiotemporal Resolution Christoph Augustin, Aurel Neic, Manfred Liebmann, Gundolf Haase, and Gernot Plank, University of Graz, Austria
Lunch Break 11:55 AM-1:30 PM Attendees on their own
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
PD1 Panel Student Careers 12:15 PM-1:30 PM
Saturday, March 14
IP2
Room:355 D
Model Reduction Trouble with Scales? 1:30 PM-2:15 PM
Sponsored by KAUST
Room:355
Chair: Evrim Acar, University of Copenhagen, Denmark
Chair: Pavel Bochev, Sandia National Laboratories, USA
Chair: William G. Kolata, SIAM, USA
Sandia National Laboratories, USA
Scientific and technological advances call for more and more complex models as well as systematic ways of complementing them by observational data. Despite the ever increasing computing capacity, ironically, the need for quantifiable model reduction concepts is also gaining increasing importance in numerous application contexts. Examples are large scale design or online optimization tasks, uncertainty quantification or inversion problems some of which may only become feasible through employing reduced models. Starting from a flow scenario with microscales this talk highlights several aspects of related model reduction strategies with particular focus on accuracy and stability guarantees, presence of small scales, singular perturbations, and high dimensionality. We address some of the key ingredients, revolving around error-residual relations, rateoptimality as a benchmark notion, adaptive or greedy methods, separation of variables. The discussion is illustrated by numerical examples.
Jeffrey Saltzman
Wolfgang Dahmen
AstraZeneca, USA
RWTH Aachen, Germany
Chair: Patrick O’Leary, Kitware, Inc., USA The panelists will present an overview careers in CSE in Academia, Industry and National Laboratories. The session will begin with brief presentations by the panelists, followed by an open discussion and question period with students in the audience. Lunch will be provided. Attendance is limited to current undergraduate and graduate students. If you did not register but would like to attend, please see a SIAM staff member at the registration desk
Thomas Grandine The Boeing Company, USA David Keyes KAUST, Saudia Arabia and Columbia University, USA
Cynthia Phillips
Intermission 2:15 PM-2:25 PM
37
Saturday, March 14
MS27 Featured Minisymposium: Fast Multipole Methods Maturing at 30 Years 2:25 PM-4:05 PM Room:355 In the last five years or so, research on fast multipole methods has been buzzing and it seems that finally this “top-10” algorithm may be reaching its potential. The mathematicians have always continued making progress and improvements, but lately we see many developments in the computer science aspects of implementing the algorithm with high performance, and the applications arena. This featured minisymposium will present leaders in the field discussing the most resent advances and giving perspectives for the future. It will complement several regular minisymposia being presented at the conference.
Organizer: Lorena A. Barba George Washington University, USA 2:25-2:45 Overview of the Field and the Community of Fast Multipole Methods Lorena A. Barba, George Washington University, USA 2:50-3:10 N-body Methods in Computational Science and Engineering George Biros, University of Texas at Austin, USA 3:15-3:35 Computer Science Aspects of Fast Multipole Methods Richard Vuduc, Georgia Institute of Technology, USA 3:40-4:00 The Geometry of the Fast Multipole Methods Lexing Ying, Stanford University, USA
38
Saturday, March 14
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS29
MS30
Geometric Methods for Graph Partitioning Part II of II 2:25 PM-4:05 PM
Adaptive Model Order Reduction - Part II of III 2:25 PM-4:05 PM Room:250 A
For Part 1 see MS2 For Part 3 see MS53 Data-driven methods are transforming the engineering, physical and biological sciences. Indeed, innovative uses of machine learning, dimensionality reduction, sparse sensing, and/or network characterization techniques are allowing for significant advances in engineering designs for the prediction and control of highly complex, often networked, systems. This minisymposium brings together experts who are integrating together one or more of the aforementioned methodologies with the goal of providing transformative analytic tools and algorithms for characterizing the underlying, lowdimensional, dynamics of the complex system.
Room:355 D
Organizer: J. Nathan Kutz
Organizer: Dominique Zosso
For Part 1 see MS4 For Part 3 see MS55 Classical model reduction follows a static approach where the reduced-order model is built once in the offline phase and then remains unchanged in the online phase. Adaptive model reduction breaks with this rigid splitting and adapts the reduced order model online. Such adaptive techniques include localization approaches, where one of several pre-computed local models is selected depending on the current state of the system; dictionary approaches, where the reduced basis is assembled on demand from pre-computed basis vectors; and updating methods, which incorporate new data or other information in the reduced-order model.
MS28 Data-methods for Complex Systems - Part II of III 2:25 PM-4:05 PM Room:355 A
University of Washington, USA
Organizer: Steven Brunton University of Washington, USA
Organizer: Joshua Proctor Institute for Disease Modeling, USA 2:25-2:45 Data-Driven Modeling of Complex Systems with Control Joshua L. Proctor, Institute for Disease Modeling, USA 2:50-3:10 Low-Complexity Stochastic Modeling of Turbulent Flows Mihailo R. Jovanovic, University of Minnesota, USA 3:15-3:35 A Deim Induced Cur Factorization Danny C. Sorensen, Rice University, USA 3:40-4:00 Data Mining and Coarse Graining for Network Evolution Problems Yannis Kevrikidis, Princeton University, USA
For Part 1 see MS3 The graph partitioning problem of finding meaningful clusters in a dataset has a variety of important applications in areas such as machine learning, image analysis, and topic modeling. A variety of new methods have recently been introduced based on ideas from geometric analysis, compressive sensing, spectral graph theory, and nonlinear partial differential equations. This minisymposium aims to bring together mathematicians and scientists working on these methods to share new results and exchange ideas.
Organizer: Braxton Osting University of Utah, USA University of California, Los Angeles, USA
Organizer: Benjamin Peherstorfer
2:25-2:45 Sampling of Dynamic Graphs and Recovery of the Spectral Properties Nathan D. Monnig, Conrad Hougen, and Francois G. Meyer, University of Colorado Boulder, USA
Massachusetts Institute of Technology, USA
2:50-3:10 Consistency of Variational Partitioning of Point Clouds Nicolas Garcia Trillos and Dejan Slepcev, Carnegie Mellon University, USA; James von Brecht, University of California, Los Angeles, USA; Thomas Laurent, Loyola Marymount University, USA; Xavier Bresson, Swiss Federal Institute of Technology, Switzerland 3:15-3:35 Geometric Methods in Image Processing, Networks, and Machine Learning Andrea L. Bertozzi, University of California, Los Angeles, USA 3:40-4:00 A Panoply of Graph-ported PDEs and Processes Yves van Gennip, Nottingham University, United Kingdom
Organizer: David Amsallem Stanford University, USA 2:25-2:45 Geometric Methods in Adaptive Model Order Reduction Ralf Zimmermann, TU Braunschweig, Germany; Thomas Franz, German Aerospace Center (DLR), Germany 2:50-3:10 An Adaptive and Efficient Greedy Procedure for the Optimal Training of Parametric ReducedOrder Models Arthur Paul-Dubois-Taine and David Amsallem, Stanford University, USA 3:15-3:35 Error Estimation for HyperReduced Elastoviscoplastic Models David Ryckelynck, Mines ParisTech, France 3:40-4:00 Symplectic Model Reduction for Hamiltonian Systems Kamran Mohseni, University of Florida, Gainesville, USA; Liqian Peng, University of Florida, USA
2015 SIAM Conference on Computational Science and Engineering
39
Saturday, March 14
Saturday, March 14
Saturday, March 14
MS31
MS32
MS33
State-of-the-art Iterative Solvers for Inverse Problems - Part II of II 2:25 PM-4:05 PM
Computational Techniques for Time Dependent Coupled Multiphysics and Multiscale Problems - Part II of III 2:25 PM-4:05 PM
Active Subspaces for Dimension Reduction: Theory and Applications - Part II of III 2:25 PM-4:05 PM
Room:250 C
For Part 1 see MS7 For Part 3 see MS58 Parameter studies like uncertainty quantification and optimization are infeasible for complex simulations with more than a handful of inputs. When a simulation contains several inputs, the engineer may seek an alternative parameterization with fewer variables to enable such studies. Active subspaces are an emerging set of tools for dimension reduction in complex models. The active subspace is a set of directions in the space of input variables that correspond to greater change, on average, in a quantity of interest. This minisymposium will explore (i) recent developments in algorithms that exploit active subspaces and (ii) applications from across engineering disciplines.
Room:250 B For Part 1 see MS5 Efficient linear solvers are a critical component in the large-scale implementations of algorithms for inverse problems. Examples include solvers for the Newton/Gauss-Newton step and solving the discretized PDEs which comprise the “forward problem”. This session will focus on advances in iterative solvers and preconditioners for linear and nonlinear inverse problems. Applications of the algorithms include diverse areas ranging from medical imaging to environmental sciences.
Room:250 D
Tufts University, USA
For Part 1 see MS6 For Part 3 see MS57 The simulation of multiphysics and multiscale systems is important for a wide range of applications including fluid structure interaction and conjugate heat transfer. Of particular interest is the formulation of solvers built by reusing codes for the various subproblems for which it is common to use ad hoc coupling or time integration schemes. This minisymposium discusses new and practical approaches to advance the state of the art, and to help bridge the gap between researchers working in the various fields. Important aspects include time adaptivity, high order methods, multirate schemes, stability, coupling conditions, and the overall solver performance.
Organizer: Eric Miller
Organizer: Jeffrey W. Banks
Colorado School of Mines, USA
Rensselaer Polytechnic Institute, USA
2:25-2:45 An Approach to Big Data in Inverse Problems Ellen B. Le, Aaron Myers, and Tan Bui-Thanh, University of Texas at Austin, USA
Organizer: Arvind Saibaba Tufts University, USA
Organizer: Misha E. Kilmer
Tufts University, USA 2:25-2:45 An Iterative Algorithm for Large-Scale Tikhonov Regularization Julianne Chung, Virginia Tech, USA; Katrina Palmer, Appalachian State University, USA 2:50-3:10 The Arnoldi-Tikhonov Framework: Choice of Regularization Parameters and Matrices Silvia Gazzola, University of Padova, Italy; James G. Nagy, Emory University, USA; Paolo Novatti, University of Padova, Italy 3:15-3:35 Flexible Krylov Subspace Methods for Shifted Systems with Multiple Right Hand Sides Tania Bakhos, Stanford University, USA; Arvind Saibaba, Tufts University, USA; Peter K. Kitanidis, Stanford University, USA 3:40-4:00 Unbiased Predictive Risk and Discrepancy Principles Applied for LSQR Solutions of Ill-posed Least Squares Rosemary A. Renaut, Arizona State University, USA
2:25-2:45 Half-Imex Time Integrators for Large Scale Simulations of Turbulent Incompressible Flows Santiago Badia, Universitat Politecnica de Catalunya, Spain; Oriol Colomes, International Center for Numerical Methods in Engineering, Spain 2:50-3:10 Overview of AddedMass Partitioned Algorithms for FSI Simulations William Henshaw, Rensselaer Polytechnic Institute, USA 3:15-3:35 Overcoming the Added Mass Instability for Coupling Incompressible Flows and Elastic Beams Longfei Li, Rensselaer Polytechnic Institute, USA 3:40-4:00 Partitioned Algorithms for FSI Problems Involving Elastic Solids Coupled to Compressible and Incompressible Fluids Donald W. Schwendeman, Rensselaer Polytechnic Institute, USA
Organizer: Paul Constantine
2:50-3:10 Likelihood-Informed Dimension Reduction for Bayesian Inverse Problems Tiangang Cui, Massachusetts Institute of Technology, USA; James R. Martin, University of Texas at Austin, USA; Youssef M. Marzouk, Massachusetts Institute of Technology, USA; Antti Solonen, Lappeenranta University of Technology, Finland; Alessio Spantini, Massachusetts Institute of Technology, USA; Luis Tenorio, Colorado School of Mines, USA 3:15-3:35 Parameter Selection Techniques for Disease Models Jared Cook, Asbury University, USA; Nicholas Myers, University of Wisconsin, Milwaukee, USA; Nina Ning, George Washington University, USA; Mami Wentworth and Ralph C. Smith, North Carolina State University, USA 3:40-4:00 Active Subspaces for the Design of Supersonic Low-Boom Aircraft Trent W. Lukaczyk, Juan J. Alonso, and Francisco Palacios, Stanford University, USA
40
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS34
MS35
MS36
Advances in Algorithms for Uncertainty Quantification in Large-scale Inverse Problems - Part II of III 2:25 PM-4:05 PM
Matrix-Free Methods for Large-Scale Optimization with Applications 2:25 PM-4:05 PM
Embedded Boundary and Interface Techniques Part II of II 2:25 PM-3:40 PM
Room:250 F
Room:251 A
Room:250 E
For Part 1 see MS9 Many applications can be formulated as large-scale optimization problems, including inverse problems, medical and seismic imaging, classification in machine learning, data assimilation in weather prediction, and matrix decompositions. For all of them, explicit modeling is prohibitive and matrixfree methods are essential for competitive performance. First-order methods have proven widely successful in recent years. However, recent developments suggest that matrix-free second-order methods, such as interior-point methods, can be competitive. The goal of the minisymposium is to explore those ideas, compare to first-order methods, and promote further research and collaboration between the optimization and applied communities in developing large-scale matrix-free methods.
For Part 1 see MS10 Many problems in science and engineering are modeled by partial differential equations (PDEs) in domains with irregular geometry and/or with interfaces. Development of special algorithms are required to restore the accuracy of the numerical schemes near arbitrary boundaries/or interfaces. Recently, much of the progress has been made for designing highly accurate and efficient numerical methods for such problems which employ only simple Cartesian grids. Developments and applications of XFEM, Immersed Interface Methods and Difference Potentials Methods, as well as open questions in the field will be discussed in details.
Saturday, March 14
For Part 1 see MS8 For Part 3 see MS59 Quantification of uncertainty in largescale inverse problems governed by partial differential equations presents significant challenges due to computationally expensive parameter-to-observable maps and highdimensional parameter spaces. This minisymposium presents recent advances in algorithms that make UQ for large-scale linear and nonlinear inverse problems tractable by exploiting problem structure.
Organizer: Georg Stadler Courant Institute of Mathematical Sciences, New York University, USA
Organizer: George Biros University of Texas at Austin, USA
Organizer: Omar Ghattas University of Texas at Austin, USA
Organizer: Youssef M. Marzouk Massachusetts Institute of Technology, USA 2:25-2:45 An Empirical Objective Bayes Method for Large Inverse Problems Peter K. Kitanidis, Stanford University, USA 2:50-3:10 Applying UQ Approaches to Random Ordinary Differential Equations Tobias Neckel, Hans-Joachim Bungartz, and Alfredo Parra, Technische Universität München, Germany 3:15-3:35 Goal-Oriented Model Adaptivity for Inference Vikram Garg, University of Texas at Austin, USA; Harriet Li and Karen E. Willcox, Massachusetts Institute of Technology, USA 3:40-4:00 Sparse Grid and Reduced Basis Approximation of Bayesian Inverse Problems Peng Chen and Christoph Schwab, ETH Zürich, Switzerland
Organizer: Dominique Orban École Polytechnique de Montréal, Canada
Organizer: Aleksandr Aravkin IBM T.J. Watson Research Center, USA 2:25-2:45 Matrix Free Quadraticpenalty Methods for PDE-constrained Optimization Bas Peters and Felix J. Herrmann, University of British Columbia, Canada 2:50-3:10 Matrix-Free Solvers for Robust PCA and Distance Matrix Completion Stephen Becker, University of Colorado Boulder, USA; Aleksandr Aravkin and Aurelie Lozano, IBM T.J. Watson Research Center, USA 3:15-3:35 Compressing Clustered Data using Sparse NMF Michael A. Saunders and San Kim, Stanford University, USA 3:40-4:00 Dimensionality Reduction and Uncertainty Quantification for Inverse Problems Tristan van Leeuwen, Utrecht University, The Netherlands
Organizer: Yekaterina Epshteyn University of Utah, USA
Organizer: Gunilla Kreiss Uppsala University, Sweden 2:25-2:45 A Fourth Order Accurate Embedded Boundary Method for the Wave Equation in Second Order Form Daniel Appelo, University of New Mexico, USA 2:50-3:10 High-Order Numerical Methods for Elliptic Interface Problems Michael Medvinsky and Yekaterina Epshteyn, University of Utah, USA; Semyon V. Tsynkov, North Carolina State University, USA; Eli Turkel, Tel Aviv University, Israel 3:15-3:35 A Nitsche Stabilized Fictitious Domain Finite Element Method for the Wave Equation Simon Sticko and Gunilla Kreiss, Uppsala University, Sweden
2015 SIAM Conference on Computational Science and Engineering
41
Saturday, March 14
Saturday, March 14
Saturday, March 14
MS37
MS38
MS39
Physics-compatible Discretization on Multiphysics Systems and Efficient Multilevel Solvers Part II of II 2:25 PM-4:05 PM
Innovative Algorithms for Eigenvalue and Singular Value Decomposition Part II of III 2:25 PM-4:05 PM
AWM Workshop - Career Development: Celebrating Firsts - Panel and Roundtable 2:25 PM-4:05 PM
Room:251 C
Room:251 D
Room:251 B
For Part 1 see MS13 Sponsored by Association for Women in Mathematics (AWM)
Organizer: Young-Ju Lee
For Part 1 see MS12 For Part 3 see MS63 The minisymposium will focuses on the first steps taken on the development of novel software methodologies and algorithm for the next generation of HPC systems. Some scale challenges will be addressed; the goal is to close the “application-architecture peak performance gap” by exploring algorithms and runtime improvements that will enable key science applications to better exploit the architectural features of the extreme-scale systems. The contributed talks will cover new approaches that can overcome the limitations of existing dense/ sparse eigensolver libraries on platforms that require fine granularity and memoryaware computational tasks combined with asynchronism in parallel execution.
Texas State University, USA
Organizer: Azzam Haidar
Trinity University, USA
2:25-2:45 Solver for StructurePreserving Discretization of Incompressible MHD Equations Yicong Ma, Pennsylvania State University, USA; Xiaozhe Hu, Tufts University, USA; Jinchao Xu, Pennsylvania State University, USA
University of Tennessee, Knoxville, USA
Organizer: Yekaterina Epshteyn
Organizer: Piotr Luszczek
University of Utah, USA
University of Tennessee, Knoxville, USA
Speakers To Be Announced
For Part 1 see MS11 Design of the appropriate discretization for the PDE systems can oftentimes be done by taking into account the detailed characteristics of the PDE models in relation to the underlying Physics. In many cases, efficient multilevel iterative solvers can be developed based on such discretization. Therefore, such discretization is typically solver-friendly. The minisymposium is to disseminate the state-of-art developments in physics-compatible discretization on multi-physics systems and the corresponding efficient multilevel solvers.
Organizer: Yunrong Zhu Idaho State University, USA
2:50-3:10 Modeling and Numerical Studies for Fluid-Structure Interactions Pengtao Sun, University of Nevada, Las Vegas, USA 3:15-3:35 Cascadic Multilevel for Saddle Point Least-Squares Methods Constantin Bacuta, University of Delaware, USA 3:40-4:00 Multigrid Method for Linear Elasticity with Weakly Imposed Symmetry Young Ju Lee, Texas State University, USA
Organizer: Stanimire Tomov University of Tennessee, Knoxville, USA 2:25-2:45 A Parallel Multishift QZ Algorithm with Aggressive Early Deflation for Distributed Memory HPC Systems Björn Adlerborn and Bo T. Kågström, Umeå University, Sweden; Daniel Kressner, EPFL, Switzerland 2:50-3:10 Algorithms for HessenbergTriangular Reduction in Parallel Björn Adlerborn, Lars Karlsson, and Bo T. Kågström, Umeå University, Sweden 3:15-3:35 Avoiding Communication in Distributed-Memory Tridiagonalization Grey Ballard, Sandia National Laboratories, USA; James W. Demmel, Nicholas Knight, and Edgar Solomonik, University of California, Berkeley, USA 3:40-4:00 Performance Evaluation of Sparse Matrix-Vector Multiplication Using GPU/MIC Cluster Hiroshi Maeda and Daisuke Takahashi, University of Tsukuba, Japan
Minisymposium speakers and panelists will provide an overview of challenges and strategies for success as firsts in their respective field, career path, or organization. We will discuss how to approach the obstacles women mathematicians and scientist face when encountering firsttime career and career-related life events, including: obtaining your first position, tenure track, work-life balance, funding, and post-tenure success. Topical discussions will be led by panelists.
Organizer: Elebeoba May University of Houston, USA
Organizer: Hoa Nguyen
42
Saturday, March 14
MS40 NSF-SIAM Symposium on Mathematical and Computational Aspects of Materials Science Part II of III 2:25 PM-4:05 PM Room:251 E For Part 1 see MS14 For Part 3 see MS65 The materials science community has embarked on various large-scale computational projects. We mention the materials genome project---a broad survey of thousands of materials---and several deepmining projects of a few selected materials. These projects offer interesting opportunities for mathematical and computational scientists to advance the state of the art in materials research by developing mathematical models, computational algorithms, and tools for data analysis. The Division of Materials Research (DMR) and Division of Mathematical Sciences (DMS) of the National Science Foundation are encouraging collaborations of mathematicians and materials scientists through several mechanisms (MSII, DMREF, etc.). The speakers in this symposium will will highlight opportunities for joint research projects.
Organizer: Hans G. Kaper Argonne National Laboratory and Georgetown University, USA 2:25-2:45 Structural Optimization and 3D Printing Robert V. Kohn, Courant Institute of Mathematical Sciences, New York University, USA 2:50-3:10 DNA-Functionalized Nanoparticle Assembly and Crystallization Monica Olvera De La Cruz, Northwestern University, USA 3:15-3:35 Mathematical Challenges in Nonequilibrium Approaches to Amorphous Solids: Quantifying Disorder, Predicting Plasticity, Accelerating Simulation Michael Falk, Johns Hopkins University, USA 3:40-4:00 Integrating Mathematical Modeling and Computer Simulation with Experimental Synthesis and Characterization of Materials Long-qing Chen, Pennsylvania State University, USA
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS41
MS42
Volume Penalty and Fourier Methods for PDEs on Irregular Domains Part I of II 2:25 PM-3:40 PM
Theory, Computation and Experiments in Biofluids Part II of II 2:25 PM-4:05 PM Room:254 A
Room:251 F
For Part 1 see MS16 Over the last two decades the study of systemic biofluids and their interactions with engineering therapeutics and implants has rapidly developed into an interdisciplinary approach of utilizing Mathematics, Bioengineering, Biomedical, and the Biological Sciences to address the complexities and challenges of this field. Eight speakers in this two part minisymposium will present their research spanning mathematical analysis, algorithm development, complex simulations of biological systems and coordination between modeling and experiments. The diversity of these presentations shows the importance of interdisciplinary research on systemic biofluids to understand its wide range of domains from basic processes of fluid dynamics to technological applications.
For Part 2 see MS66 Volume penalty methods allow one to extend the domain that a partial differential equation is defined on (ie. an irregular geometry) to that of a simple rectangular domain through the addition of a volume penalty, forcing term. The resulting penalized equations then provide a flexible and efficient framework for solving (moving) interface and boundary problems via a solution using I) equispaced grids, ii) Nonconforming FEM, iii) Fourier methods. This minisymposium brings together mathematicians and practitioners to discuss both theoretical aspects such as convergence rates and stability, as well as applications and simulations of physical phenomena.
Organizer: David Shirokoff New Jersey Institute of Technology, USA 2:25-2:45 A New Penalization Method for the Shallow Water Equations with Applications to Global Ocean Flow Nicholas Kevlahan, McMaster University, Canada 2:50-3:10 Imposing Dirichlet and Neumann Conditions in Fourier Pseudospectral Methods Using Volume Penalization Kai Schneider, Aix-Marseille Université, France 3:15-3:35 High-Order Fourier-Penalty Methods for PDEs on Irregular Domains David Shirokoff, New Jersey Institute of Technology, USA
Organizer: Hoa Nguyen Trinity University, USA
Organizer: John Carrola Southwest Research Institute, USA 2:25-2:45 The Effect of Curved or Flat Edges Microchannels on Vortex Entrapment of Particles as seen in Lattice-Boltzmann Simulations John Carrola and Hakan Basagaoglu, Southwest Research Institute, USA 2:50-3:10 Scaffold-free Threedimensional Hepatocyte Assembly for Liver Tissue Engineering Utkan Demirci and Pu Chen, Stanford University, USA 3:15-3:35 Brownian Motion of Arbitrarily Shaped Particles Confined in Two-Dimensions Qi-Huo Wei, Kent State University, USA 3:40-4:00 Designing Self-Propelling Microgel Swimmer Alexander Alexeev, Svetoslav Nikolov, and Peter Yeh, Georgia Institute of Technology, USA
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
MS43 Parallel Programming Models, Algorithms and Frameworks for Extreme Computing - Part II of III 2:25 PM-4:05 PM Room:254 B For Part 1 see MS17 For Part 3 see MS68 Multicore/manycore processors and accelerators are universally available as both collections of homogeneous standard microprocessors and as attached heterogeneous co-processors. Application and library software developers may often effectively use these processors and some general approaches have emerged. It is widely recognized that careful design of software and data structures, with effective memory management, are the most critical issues to obtain scalable optimized performance on those systems. In these minisymposia we discuss current experiences and development of applications, libraries and frameworks using a variety of hardware. Speakers will address performance results and software design.
Organizer: Kengo Nakajima University of Tokyo, Japan
Organizer: Michael Heroux Sandia National Laboratories, USA
3:15-3:35 Optimization of Preconditioned Iterative Linear Solvers Using Openmp/openacc on Gpu and Mic Satoshi Ohshima, Masaharu Matsumoto, Takahiro Katagiri, Toshihiro Hanawa, and Kengo Nakajima, University of Tokyo, Japan 3:40-4:00 A Kokkos Implementation of Albany: A Performance Portable Multiphysics Simulation Code Irina Demeshko, H. Carter Edwards, Michael Heroux, Roger P. Pawlowski, Eric Phipps, and Andrew Salinger, Sandia National Laboratories, USA
43
Saturday, March 14
MS44 Advanced Algorithms in Computational Electromagnetics Part II of II 2:25 PM-4:05 PM Room:254 C For Part 1 see MS18 The minisymposium focuses on advanced numerical algorithms and high performance simulations for electromagnetic systems, including photonics crystals, plasmonics, solar cells, and quantum electronic devices. Main topics include high-order PDE/ODE discretizations, Green function, integral equation, boundary element method, and density matrix approaches. Scalable algorithms and efficient preconditioning strategies are also discussed for solving large-scale systems on the advanced computing platforms.
Organizer: David P. Nicholls University of Illinois, Chicago, USA
Organizer: MiSun Min Argonne National Laboratory, USA 2:25-2:45 Resonances of a Finite OneDimensional Photonic Crystal with a Defect Fadil Santosa, University of Minnesota, USA
CNRS/LIFL and INRIA, France
2:50-3:10 Electromagnetic Field Enhancement for Metallic Nano-gaps Junshan Lin, Auburn University, USA
2:25-2:45 OCCA: An Extensible Portability Layer for Many-Core Programming Tim Warburton and David Medina, Rice University, USA; Amik St-Cyr, Shell International Exploration & Production B.V., Netherlands
3:15-3:35 An Efficient Spectral Element Helmholtz Solver with an Accurate Treatment for Transparent Boundary Condition for Periodic Lossy Media Ying He, University of California, Davis, USA
2:50-3:10 Evaluations of Directive Based Programming Model for GPUs and Extensions for Performance Portability Tetsuya Hoshino, Tokyo Institute of Technology, Japan; Naoya Maruyama, RIKEN, Japan; Satoshi Matsuoka, Tokyo Institute of Technology, Japan
3:40-4:00 A High Order Perturbation of Surfaces Method for Simulating Surface Plasmons on Periodic Gratings David P. Nicholls, University of Illinois, Chicago, USA
Organizer: Serge G. Petiton
continued in next column
44
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
MS45 Computational Methods for Kinetic Equations and Related Models - Part II of V 2:25 PM-4:05 PM Room:150 AB For Part 1 see MS19 For Part 3 see MS70 Kinetic descriptions play an important role in a variety of physical, biological, and even social applications. Unfortunately, the large phase space associated with the kinetic description has in the past made simulations impractical in most settings. However, recent advances in computer resources and numerical algorithms are making kinetic models more tractable, and this trend is expected to continue in the future. The purpose of this minisymposium is to report on the continuing progress on numerical analysis and computational science for kinetic equations. It brings together researchers from applied mathematics, computational science, physics, and engineering communities to discuss their work and exchange ideas.
Organizer: Martin Frank RWTH - Aachen University of Technology, Germany
Organizer: Cory Hauck Oak Ridge National Laboratory, USA
Organizer: Ryan G. McClarren Texas A&M University, USA
Organizer: Jingmei Qiu University of Houston, USA
Organizer: Jeffrey A. Willert Los Alamos National Laboratory, USA 2:25-2:45 Performance of Parallel Algorithms for Particle Transport on Massively Parallel Architectures Marvin L. Adams, Michael Adams, W. Daryl Hawkins, Timmie Smith, Lawrence Rauchwerger, and Nancy Amato, Texas A&M University, USA; Teresa S. Bailey and Robert Falgout, Lawrence Livermore National Laboratory, USA
continued in next column
2:50-3:10 Particle-Particle, ParticleMesh Methods for Electromagnetic Problems Andrew J. Christlieb and Eric Wolf, Michigan State University, USA 3:15-3:35 Uncertainty Quantification in Kinetic Theory Shi Jin, Shanghai Jiao Tong University, China, and the University of WisconsinMadison, USA; Dongbin Xiu and Xueyu Zhu, University of Utah, USA 3:40-4:00 Stochastic Galerkin Method for Hamilton-Jacobi Equations with Uncertainty Jingwei Hu, Purdue University, USA; Shi Jin, Shanghai Jiao Tong University, China, and the University of WisconsinMadison, USA; Dongbin Xiu, University of Utah, USA
Saturday, March 14
MS46 Advances in Large-scale Forward and Inverse Ice Sheet Modeling - Part I of II 2:25 PM-4:05 PM Room:150 DE For Part 2 see MS71 Model-based projections of the dynamics of the polar ice sheets play a central role in anticipating future sea level rise. However, a number of mathematical and computational challenges place significant barriers on improving predictability of these models. These include complex and very high-aspect ratio geometries, highly nonlinear and anisotropic rheology, extremely ill-conditioned (non)linear systems, broad range of length scales, and unknown model parameters that must be inferred from heterogeneous observations, leading to an ill-posed inverse problem and the need to quantify uncertainties in its solution. Speakers in this minisymposium will address these challenges and present recent developments aimed at overcoming them.
Organizer: Noemi Petra University of California, Merced, USA
Organizer: Omar Ghattas University of Texas at Austin, USA
Organizer: Irina Kalashnikova Sandia National Laboratories, USA
Organizer: Georg Stadler Courant Institute of Mathematical Sciences, New York University, USA 2:25-2:45 Improving Grounding Line Discretization using an EmbeddedBoundary Approach in BISICLES Daniel Martin, Peter O. Schwartz, and Esmond G. Ng, Lawrence Berkeley National Laboratory, USA 2:50-3:10 On the Development and Performance of a First Order Stokes Finite Element Ice Sheet Dynamical Core Built Using Trilinos Software Components Irina Kalashnikova, Andrew Salinger, Mauro Perego, and Ray S. Tuminaro, Sandia National Laboratories, USA
continued on next page
2015 SIAM Conference on Computational Science and Engineering
3:15-3:35 A Finite Element ThreeDimensional Stokes Ice Sheet Dynamics Model with Enhanced Local Mass Conservation Wei Leng, Chinese Academy of Sciences, China; Lili Ju, University of South Carolina, USA; Max Gunzburger, Florida State University, USA 3:40-4:00 Testing the Multilayer Shallow Shelf Approximation Against Higher-order Models Guillaume Jouvet, ETH Zürich, Switzerland
Saturday, March 14
MS47 Career Fair: Alternatives to Academia - Part II of III 2:25 PM-4:05 PM Room:255 For Part 1 see MS21 For Part 3 see MS72 The career fair will feature representatives from nonacademic employers from industry and government. These representatives will be prepared to discuss with you the opportunities for internships, postdoctoral appointments and full-time jobs at their organizations.
Organizer: William G. Kolata SIAM, USA
Organizer: Kristin O’Neill SIAM, USA This is the most current list at time fo printing. • Argonne National Laboratory • Boeing • Hewlett-Packard Laboratories • Kitware • Lawrence Berkeley National Laboratory • Lawrence Livermore National Laboratory • MathWorks • MIT Lincoln Laboratory • National Institute of Standards & Technology • NSA • NVIDIA • Oak Ridge National Laboratory • Quantlab The most current list of participating companies is available at www.siam. org/meetings/cse15/career.php.
45
Saturday, March 14
MS48 Efficient and Accurate Solution Techniques for Variable Coefficient Elliptic Partial Differential Equations Part I of II 2:25 PM-4:05 PM Room:260 A For Part 2 see MS73 Efficient and accurate numerical methods for solving variable coefficient partial differential equations are important for many applications such as seismic imaging and metamaterial design. A variety of techniques including finite element, discontinuous Galerkin, and integral equation methods have been developed to tackled these challenging problems. This session brings together researchers from a broad range of research communities in an effort to build an understanding of the different techniques and open problems in the field.
Organizer: Adrianna Gillman Rice University, USA
Organizer: Lise-Marie Imbert-Gérard Courant Institute of Mathematical Sciences, New York University, USA 2:25-2:45 Yee Scheme Coupled with Linear Current in Magnetic Plasmas with Varying Coefficients Bruno Despres, University of Paris VI, France; Martin Campos Pinto and Stéphane Heuraux, CNRS, France; Filipe da Silva, Instituto Superior Tecnico, Portugal 2:50-3:10 Discontinuous Enrichment Method for Problems with Variable Coefficients Radek Tezaur and Charbel Farhat, Stanford University, USA; Irina Kalashnikova, Sandia National Laboratories, USA 3:15-3:35 Approximation of Degenerate Elliptic Equations with Muckenhoupt Coefficients: a priori and a posteriori Analyses and Efficient Solvers Abner J. Salgado, University of Tennessee, USA 3:40-4:00 Application of a Speed Up Fast Direct Solver for the Solution of the Lippmann-Schwinger Equation Carlos C. Borges, Lise-Marie Imbert-Gerard, and Sivaram Ambikasaran, Courant Institute of Mathematical Sciences, New York University, USA; Leslie Greengard, Simons Foundation and Courant Institute of Mathematical Sciences, New York University, USA
46
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS50
MS51
Recent Developments of DG/WENO Methods for Partial Differential Equations - Part II of II 2:25 PM-4:05 PM
Advances in Radial Basis Function and Other Meshfree Methods Part II of V 2:25 PM-4:05 PM
Challenges in Computational Cardiac Electrophysiology Part II of II 2:25 PM-4:05 PM
Room:260B
Room:151 AB
Room:151 G
For Part 1 see MS23 This minisymposium is to bring people together to discuss the recent advances and exchange ideas in the algorithm design of discontinuous Galerkin (DG) method and weighted essentially nonoscillatory (WENO) method, including the implementation, numerical analysis of those high-order numerical methods for solving partial differential equations. In the minisymposium, the speakers will apply those high-order numerical methods to computational fluid, biology and physics, etc. This minisymposium is a good opportunity for people to discuss with researchers from different areas, and explore more applications and future research collaborations.
For Part 1 see MS24 For Part 3 see MS75 Meshfree methods have many inherent properties that make them useful for a variety of problems in science and engineering, from fitting data to numerically solving differential equations. Such methods offer geometric flexibility, non-uniform resolution, and advantageous trade-offs between accuracy and computational costs. This minisymposium focuses primarily on methods based on radial basis functions and other more general kernels. The talks will address recent advances in the application of these methods to large-scale problems in biology, geophysics, image processing, and finance, as well as theoretical advances in the methods themselves.
For Part 1 see MS26 The use of computer simulation to model cardiac electrophysiology, from the cellular level to the whole-organ, has the potential to revolutionise mechanistic understanding, diagnosis and treatment of arrhythmias. These models are often complex and computationally demanding, and integrating the available electrical and imaging data with its inherent uncertainty makes direct clinical utility particularly challenging. This minisymposium highlights the latest advancements and new approaches to modelling cardiac electrophysiology across multiple scales, including imaging and data integration, which will pave the way for developing clinically tractable simulation environments to aid patient treatment in the future.
Saturday, March 14
MS49
Organizer: Juan Cheng Institute of Applied Physics and Computational Mathematics, China
Organizer: Varun Shankar University of Utah, USA
Organizer: Chris Cantwell
Organizer: Yang Yang
Organizer: Grady B. Wright
Imperial College London, United Kingdom
Michigan Technological University, USA
Boise State University, USA
Organizer: Frank B. Sachse
2:25-2:45 Runge-Kutta Discontinuous Galerkin Methods for the Relativistic Vlasov-Maxwell System He Yang and Fengyan Li, Rensselaer Polytechnic Institute, USA
2:25-2:45 Quadrature on Spheres and Other Manifolds Based on Kernels Joseph Ward, Texas A&M University, USA
University of Utah, USA
2:50-3:10 Solving PDEs on the Sphere via Novel Galerkin Method using Highly Localized Kernel Bases Francis J. Narcowich, Joseph Ward, and Stephen Rowe, Texas A&M University, USA
2:25-2:45 Multi-Scale Modeling in Cardiac Electrophysiology: What Are the Challenges in Front of Us? Zhilin Qu, University of California, Los Angeles, USA
2:50-3:10 Superconvergence of Discontinuous Galerkin Methods for Hyperbolic Equations in Two Space Dimensions Waixiang Cao, Beijing Computational Science Research Center, China; Yang Yang, Michigan Technological University, USA; Zhimin Zhang, Wayne State University, USA; Chi-Wang Shu, Brown University, USA 3:15-3:35 A Simple DG Scheme for Acoustic Wave Equations with Curved Interfaces and Boundaries Xiangxiong Zhang, Purdue University, USA 3:40-4:00 Runge-Kutta Discontinuous Galerkin Method with a Simple and Compact Hermit Weno Limiter Jun Zhu, Nanjing University of Aeronautics and Astronautics, China; Xinghui Zhong, Michigan State University, USA; Chi-Wang Shu, Brown University, USA; Jianxian Qiu, Xiamen University, China
3:15-3:35 Compact Scattered RBF-FD Stencils for PDEs on Surfaces Erik Lehto, Royal Institute of Technology, Stockholm, Sweden; Varun Shankar, University of Utah, USA; Grady B. Wright, Boise State University, USA
Organizer: Richard Clayton University of Sheffield, United Kingdom
2:50-3:10 Three-Dimensional Modeling of Ca2+ Signaling in Healthy and Failing Cardiomyocytes Peter Kekenes-Huskey, University of Kentucky, USA
3:40-4:00 A Least Squares-RBF Approach to Transport Problems on Surfaces Daryl J. Springer, Arizona State University, USA
continued on next page
2015 SIAM Conference on Computational Science and Engineering
3:15-3:35 High-Order Finite Element Methods for Cardiac Electrophysiology Chris Cantwell, Imperial College London, United Kingdom; Sergey B. Yakovlev, University of Utah, USA; Rheeda Ali and Nicholas Peters, Imperial College London, United Kingdom; Mike Kirby, University of Utah, USA; Spencer Sherwin, Imperial College London, United Kingdom 3:40-4:00 Assessing the Credibility of Computational Models of Cardiac Electrophysiology Pras Pathmanathan and Richard Gray, U.S. Food and Drug Administration, USA
Coffee Break 4:05 PM-4:35 PM Room:255
47
Saturday, March 14
Saturday, March 14
MS52
MS53
Featured Minisymposium: Modeling and Computing Complex Flows 4:35 PM-6:15 PM
Data-methods for Complex Systems - Part III of III 4:35 PM-6:15 PM
Room:355 Computation of unsteady multifluid flows is one of the more challenging problems in computational science. Nevertheless, considerable progress has been made in the development of numerical methods to follow the convoluted motion of complex interfaces separating different fluid phases, for a broad range of governing parameters. These successes are making it possible to pursue even more advanced problems, including additional physics and even large range of scales, as well as making it urgent to develop sophisticated models that take advance of the new data. In this minisymposium we will explore the state of the art and discuss future challenges.
For Part 2 see MS28 Data-driven methods are transforming the engineering, physical and biological sciences. Indeed, innovative uses of machine learning, dimensionality reduction, sparse sensing, and/or network characterization techniques are allowing for significant advances in engineering designs for the prediction and control of highly complex, often networked, systems. This minisymposium brings together experts who are integrating together one or more of the aforementioned methodologies with the goal of providing transformative analytic tools and algorithms for characterizing the underlying, low-dimensional, dynamics of the complex system.
Organizer: Gretar Tryggvason
Organizer: J. Nathan Kutz
Room:355 A
University of Notre Dame, USA
University of Washington, USA
4:35-4:55 Direct Numerical Simulations of Multiphase Flow: Now What? Gretar Tryggvason, University of Notre Dame, USA
Organizer: Steven Brunton
5:00-5:20 Modeling and Simulation of Multimaterial Compressible Flows Marianne M. Francois, Los Alamos National Laboratory, USA 5:25-5:45 Methods for Computing Turbulent Phase Interface Dynamics Across Multiple Scales Marcus Herrmann, Arizona State University, USA 5:50-6:10 Conservative and Accurate Geometric Transport Methods for Discontinuous Variables in Turbulent Multi-physics Two-phase Flows Olivier Desjardins, Cornell University, USA
University of Washington, USA
Organizer: Joshua Proctor Institute for Disease Modeling, USA 4:35-4:55 Discovering Underlying Nonlinear Dynamics of Complex Systems from Data Steven Brunton, University of Washington, USA 5:00-5:20 Common Manifold Learning Using Alternating Diffusion for Multimodal Signal Processing Ronald Coifman and Roy Lederman, Yale University, USA; Ronen Talmon, Technion Israel Institute of Technology, Israel 5:25-5:45 Title Not Available at Time of Publication Surya Ganguli, Stanford University, USA 5:50-6:10 Data-Driven Model Reduction to Support Decision Making in Complex Systems Karen E. Willcox, Massachusetts Institute of Technology, USA
48
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS55
MS56
Computationally Intensive Signature Discovery 4:35 PM-6:15 PM
Adaptive Model Order Reduction - Part III of III 4:35 PM-6:15 PM
Room:355 D
Room:250 A
A signature is a unique or distinguishing measurement, pattern, or collection of data that predicts, detects, or identifies a phenomenon of interest. Signature discovery challenges are commonplace in numerous domains, and they typically involve the following steps: hypothesis generation, identifying observables, specifying the measurement technique, data collection, storage, and manipulation, feature extraction, classification, and validation. The process often requires complex computational tools and methods for data processing and analysis. In this session, we will provide an overview of signature discovery, discuss relevant mathematical and computational tools, and examine signature applications in medical imaging, insect neurology, and social/organizational networks.
For Part 2 see MS30 Classical model reduction follows a static approach where the reduced-order model is built once in the offline phase and then remains unchanged in the online phase. Adaptive model reduction breaks with this rigid splitting and adapts the reduced order model online. Such adaptive techniques include localization approaches, where one of several pre-computed local models is selected depending on the current state of the system; dictionary approaches, where the reduced basis is assembled on demand from pre-computed basis vectors; and updating methods, which incorporate new data or other information in the reduced-order model.
Efficient Methods for Uncertainty Quantification by Means of Tensor Format Representations 4:35 PM-6:15 PM
Organizer: Landon H. Sego
Organizer: David Amsallem
Saturday, March 14
MS54
Pacific Northwest National Laboratory, USA 4:35-4:55 Computational Tools and Methods for Signature Discovery (Session Overview) Landon H. Sego, Pacific Northwest National Laboratory, USA 5:00-5:20 Statistics, Learning, and Optimization for Data Analysis and Visualization Ross Whitaker, University of Utah, USA 5:25-5:45 Computational Analysis of Ensemble Neural Data Recorded From An Insect Brain Debajit Saha, Chao Li, and Barani Raman, Washington University in St. Louis, USA 5:50-6:10 PhySense: Social and Organizational Network Activity Simulation for Signature Generation and Extraction Vikram Jandhyala and Arun Sathanur, University of Washington, USA
Organizer: Benjamin Peherstorfer Massachusetts Institute of Technology, USA Stanford University, USA
Room:250 B In this minisymposium we introduce new effective tensor methods for quantification of uncertainties, analysis of high dimensional data, which may come from, e.g. stochastic or multi- parametric PDEs. Tensor format representations can be used for fast (with almost linear complexity) computing different statistics and values of interest, such as mean, covariance, exceedance probabilities, confidence intervals, sensitivity indices and cumulative distribution function. A very important issue to discuss is inexact recursive iteration schemes as well as tensor rank truncation. A special attention will be devoted to the non-intrusive implementation of tensor methods for uncertainty quantification.
Organizer: Alexander Litvinenko King Abdullah University of Science & Technology (KAUST), Saudi Arabia
4:35-4:55 Numerical Study of Local Reduced Basis with Adaptive Training for Incompressible Navier-Stokes Flows Yuqi Wu and Ulrich Hetmaniuk, University of Washington, USA
Organizer: Mike Espig
5:00-5:20 Real-Time Data-to-Decision Using Adaptive Surrogate Modeling Strategies Laura Mainini and Karen E. Willcox, Massachusetts Institute of Technology, USA
4:35-4:55 Tensor Format Representations and Optimal Model Reduction for Uncertainty Quantication Mike Espig, RWTH - Aachen University of Technology, Germany
5:25-5:45 An Occam’s Razor Strategy for Field Estimation from WallMounted Sensors Kevin Kasper, ENS Cachan, France; Lionel Mathelin, CNRS, France; Mohamed Abbas-Turki and Hisham Abou-Kandil, ENS Cachan, France 5:50-6:10 Thermal Reduced Order Model Adaptation to Aero-ThermoStructural Interactions Andrew Matney and Marc P. Mignolet, Arizona State University, USA
RWTH - Aachen University of Technology, Germany
Organizer: Hermann Matthies TU Braunschweig, Germany
5:00-5:20 High-Dimensional Tensor Sampling Lars Grasedyck, RWTH - Aachen University of Technology, Germany; Jonas Ballani, EPFL, France; Melanie Kluge, RWTH Aachen University, Germany 5:25-5:45 Novel Tensor-Product Representations for Uncertaintiy Quantification Reinhold Schneider, Technische Universität Berlin, Germany 5:50-6:10 Hierarchical Tensor Approximation of ParameterDependent PDEs Jonas Ballani, EPFL, France; Lars Grasedyck, RWTH - Aachen University of Technology, Germany; Daniel Kressner, EPFL, Switzerland
2015 SIAM Conference on Computational Science and Engineering
49
Saturday, March 14
Saturday, March 14
Saturday, March 14
MS57
MS58
MS59
Computational Techniques for Time Dependent Coupled Multiphysics and Multiscale Problems Part III of III 4:35 PM-6:15 PM
Active Subspaces for Dimension Reduction: Theory and Applications Part III of III 4:35 PM-6:15 PM
Advances in Algorithms for Uncertainty Quantification in Large-scale Inverse Problems - Part III of III 4:35 PM-6:15 PM
Room:250 D
Room:250 E
Room:250 C
For Part 2 see MS33 Parameter studies like uncertainty quantification and optimization are infeasible for complex simulations with more than a handful of inputs. When a simulation contains several inputs, the engineer may seek an alternative parameterization with fewer variables to enable such studies. Active subspaces are an emerging set of tools for dimension reduction in complex models. The active subspace is a set of directions in the space of input variables that correspond to greater change, on average, in a quantity of interest. This minisymposium will explore (i) recent developments in algorithms that exploit active subspaces and (ii) applications from across engineering disciplines.
For Part 2 see MS34 Quantification of uncertainty in largescale inverse problems governed by partial differential equations presents significant challenges due to computationally expensive parameter-to-observable maps and highdimensional parameter spaces. This minisymposium presents recent advances in algorithms that make UQ for large-scale linear and nonlinear inverse problems tractable by exploiting problem structure.
Organizer: Paul Constantine
Massachusetts Institute of Technology, USA
For Part 2 see MS32 The simulation of multiphysics and multiscale systems is important for a wide range of applications including fluid structure interaction and conjugate heat transfer. Of particular interest is the formulation of solvers built by reusing codes for the various subproblems for which it is common to use ad hoc coupling or time integration schemes. This minisymposium discusses new and practical approaches to advance the state of the art, and to help bridge the gap between researchers working in the various fields. Important aspects include time adaptivity, high order methods, multirate schemes, stability, extrapolation, coupling conditions, and the overall solver performance.
Organizer: Jeffrey W. Banks
Colorado School of Mines, USA
Rensselaer Polytechnic Institute, USA
4:35-4:55 Influence of Surface and Subsurface Parameter Uncertainty and Sensitivity on the Latent Heat Flux Using An Integrated Hydrologic Model Jennifer Jefferson and Reed M. Maxwell, Colorado School of Mines, USA
Organizer: Philipp Birken Lund University, Sweden 4:35-4:55 Fluid-composit Structure Interaction Suncica Canic, University of Houston, USA; Martina Bukac, University of Notre Dame, USA; Boris Muha, University of Zagreb, Croatia 5:00-5:20 Fractional Modeling of Brain Aneurysms Yue Yu, Lehigh University, USA; George E. Karniadakis, Brown University, USA 5:25-5:45 A Tetrahedral Method for Transient Nonlinear Dynamics Computations in Solids, Fluids and Coupled Fluid Structure Problems Guglielmo Scovazzi and Xianyi Zeng, Duke University, USA; Brian Carnes and David Hensinger, Sandia National Laboratories, USA 5:50-6:10 Second Order Embedded Boundary Methods for Fluid-Structure Interaction Alex Main and Charbel Farhat, Stanford University, USA
5:00-5:20 Dimension Reduction in MCMC using Active Subspaces Carson Kent and Paul Constantine, Colorado School of Mines, USA 5:25-5:45 Exploiting Active Subspaces to Quantify Uncertainty in the Numerical Simulation of the HyShot II Scramjet Michael A. Emory and Gianluca Iaccarino, Stanford University, USA; Johan Larsson, University of Maryland, USA 5:50-6:10 Discovering An Active Subspace in a Single-Diode Solar Cell Model Mark Campanelli, National Renewable Energy Laboratory, USA; Brian Zaharatos, Colorado School of Mines, USA
Organizer: George Biros University of Texas at Austin, USA
Organizer: Omar Ghattas University of Texas at Austin, USA
Organizer: Youssef M. Marzouk Organizer: Georg Stadler Courant Institute of Mathematical Sciences, New York University, USA 4:35-4:55 Mapped Stochastic Newton Sampling Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA; Omar Ghattas, University of Texas at Austin, USA; Noemi Petra, University of California, Merced, USA 5:00-5:20 Operator Weighted MCMC on Function Spaces Tiangang Cui, Massachusetts Institute of Technology, USA; Kody Law, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Youssef M. Marzouk, Massachusetts Institute of Technology, USA 5:25-5:45 Regularising Ensemble Kalman Methods for PDE-Constrained Inverse Problems Marco Iglesias, University of Nottingham, United Kingdom 5:50-6:10 Estimation of Parameters of Chaotic Dynamic Systems Heikki Haario, Lappeenranta University of Technology, Finland
50
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS60
MS61
MS62
Surrogate Global and Integer Optimization for Computationally Expensive Simulations 4:35 PM-6:15 PM
Recent Advances in Nuclear Quantum Manybody Computation 4:35 PM-6:15 PM
Theory Implementation and Applications of HDG Methods - Part I of II 4:35 PM-5:50 PM
Room:251 A
Room:251 B
Room:250 F
The nuclear quantum many-body problem is a fundamental problem in nuclear physics. Hurdles in solving this problem include a very strong interaction, three-nucleon interactions, and complicated collective motion dynamics. In recent years, significant progress has been made to overcome these difficulties. In particular, the configuration interaction (CI) method, which requires solving a large-scale eigenvalue problem, has become very efficient on modern high performance computers. New optimization algorithms have been developed to fine tune model parameters and to extrapolate computational results. These new techniques will be described and discussed in this minisymposium.
For Part 2 see MS87 In this minisymposium, we will discuss the latest advancements related to the hybrid discontinuous Galerkin (HDG) method. HDG applies a static condensation technique within the DG framework, so that the only globally coupled degrees of freedom are those located on the mesh skeleton or trace space, greatly reducing the global system size. The method is also promising for its applicability to current and emerging parallel architectures. The HDG method has proven to be a popular method and has been applied to a variety of problems such as, steady-state diffusion, Maxwell’s equations, convectiondiffusion problems, elastodynamics, Stokes and incompressible Navier-Stokes equations.
Organizer: Meiyue Shao
Organizer: Hari Sundar
Lawrence Berkeley National Laboratory, USA
University of Utah, USA
4:35-4:55 Add, Multiply, Divide and Conquer: On-the-fly Algorithms for Many-body Calculations Calvin W. Johnson, San Diego State University, USA
University of Texas at Austin, USA
Saturday, March 14
Optimization applications in STEM disciplines often require evaluating a computationally expensive simulation-based objective function (several minutes to hours/ simulation). An analytical description of the objective function, its derivatives, and the number of local minima are not available (black-box) and can thus not be exploited by the optimization algorithm. Surrogate models have been developed for continuous optimization to cheaply approximate the objective function and efficiently find the global minimum. However, few surrogate model algorithms exist that are able to address the following: integer variables, incorporation of partial information, and/or efficient parallel computation, all of which will be discussed in this minisymposium.
Organizer: Juliane Mueller Lawrence Berkeley National Laboratory, USA
Organizer: David Bindel Cornell University, USA
Organizer: Christine A. Shoemaker Cornell University, USA 4:35-4:55 Miso: Mixed-Integer Surrogate Optimization for Computationally Expensive BlackBox Problems Juliane Mueller, Lawrence Berkeley National Laboratory, USA 5:00-5:20 RBF Response Surfaces with Inequality Constraints David Bindel, Cornell University, USA 5:25-5:45 Efficient Multi-Start for Global Optimization in Accelerator Design Jeffrey M. Larson and Stefan Wild, Argonne National Laboratory, USA 5:50-6:10 Parallel Surrogate Global Optimization with Pareto Centers for Single Objective Expensive Functions Christine A. Shoemaker, Tipaluck Krityakierne, and Taimoor Akhtar, Cornell University, USA
5:00-5:20 Derivative-free Optimization Techniques in ab initio Nuclear Structure Calculations Masha Sosonkina, Old Dominion University, USA 5:25-5:45 Symmetry-adapted No-core Shell Model for First Principle Lage Scale Computations of Atomic Nuclei Tomas Dytrych, Louisiana State University, USA 5:50-6:10 Multi-Level LOBPCG Method in MFDn Meiyue Shao, Lawrence Berkeley National Laboratory, USA
Organizer: Tan Bui-Thanh Organizer: Cuong Nguyen Massachusetts Institute of Technology, USA 4:35-4:55 Parallel hp-multigrid for HDG Tan Bui, University of Texas at Austin, USA; Hari Sundar, University of Utah, USA 5:00-5:20 HDG Methods for the p-Laplacian Bernardo Cockburn, University of Minnesota, Minneapolis, USA; Jiguang Shen, University of Minnesota, USA 5:25-5:45 HDG Method for Linear Elasticity Weifeng Qiu, City University of Hong Kong, Hong Kong
2015 SIAM Conference on Computational Science and Engineering
51
Saturday, March 14
Saturday, March 14
Saturday, March 14
MS63
MS64
MS65
Innovative Algorithms for Eigenvalue and Singular Value Decomposition Part III of III 4:35 PM-6:15 PM
Efficient High-order Numerical Methods for Nonlinear PDEs - Part II of II 4:35 PM-6:15 PM
Room:251 C
For Part 1 see MS15 Recent years have seen growing trends in the development of efficient highorder numerical schemes for nonlinear PDEs including hyperbolic equations, Hamilton-Jacobi equations, and quantum and kinetic models. Many challenges in computational efficiency, accuracy, multiscales, and theoretical foundations are actively addressed for various applications. This minisymposium will bring together researchers to exchange ideas and recent developments in these fields, with applications such as rarefied gas dynamics, semi-conductor simulation, optimal transportation and optimal control.
NSF-SIAM Symposium on Mathematical and Computational Aspects of Materials Science Part III of III 4:35 PM-6:15 PM
For Part 2 see MS38 The minisymposium will focuses on the first steps taken on the development of novel software methodologies and algorithm for the next generation of HPC systems. Some scale challenges will be addressed; the goal is to close the “application-architecture peak performance gap” by exploring algorithms and runtime improvements that will enable key science applications to better exploit the architectural features of the extreme-scale systems. The contributed talks will cover new approaches that can overcome the limitations of existing dense/ sparse eigensolver libraries on platforms that require fine granularity and memoryaware computational tasks combined with asynchronism in parallel execution.
Organizer: Azzam Haidar University of Tennessee, Knoxville, USA
Organizer: Piotr Luszczek University of Tennessee, Knoxville, USA
Organizer: Stanimire Tomov University of Tennessee, Knoxville, USA 4:35-4:55 Revisiting SVD(A) through EIG(T) for Sca/LAPACK Osni A. Marques, Lawrence Berkeley National Laboratory, USA 5:00-5:20 Solving a Parameterized Eigenvalue Problem from Regularized Total Least Squares Jesse L. Barlow, Geunseop Lee, and Haoying Fu, Pennsylvania State University, USA 5:25-5:45 Performance Evaluation of EigenExa Dense Eigensolver on the Oakleaf-Fx Supercomputer System Takeshi Fukaya and Toshiyuki Imamura, RIKEN, Japan 5:50-6:10 Scaling Comparison of Dense Eigensolvers and Purification Techniques to Large Node Counts Xing Liu and Edmond Chow, Georgia Institute of Technology, USA
Room:251 D
Organizer: Yingda Cheng Michigan State University, USA
Organizer: Fengyan Li Rensselaer Polytechnic Institute, USA 4:35-4:55 High Order Methods for Traveltime and Amplitude in Geometrical Optics Songting Luo, Iowa State University, USA; Jianliang Qian, Michigan State University, USA; Robert Burridge, University of New Mexico, USA 5:00-5:20 High-Order Gas Evolution Model for Computational Fluid Dynamics Kun Xu, University of Science & Technology, Hong Kong 5:25-5:45 An Efficient Spectral Method for the Euler-Lagrange Equations of Minimum Action Methods Haijun Yu, Institute of Computational Mathematics, China; Xiaoliang Wan, Louisiana State University, USA 5:50-6:10 Higher-Order Filtered Methods for Nonlinear Partial Differential Equations Brittany Froese, University of Texas at Austin, USA; Adam M. Oberman, McGill University, Canada
Room:251 E For Part 2 see MS40 The materials science community has embarked on various large-scale computational projects. We mention the materials genome project---a broad survey of thousands of materials---and several deep-mining projects of a few selected materials. These projects offer interesting opportunities for mathematical and computational scientists to advance the state of the art in materials research by developing mathematical models, computational algorithms, and tools for data analysis. The Division of Materials Research (DMR) and Division of Mathematical Sciences (DMS) of the National Science Foundation are encouraging collaborations of mathematicians and materials scientists through several mechanisms (MSII, DMREF, etc.). The speakers in this symposium will will highlight opportunities for joint research projects.
Organizer: Hans G. Kaper Argonne National Laboratory and Georgetown University, USA 4:35-4:55 Materials from Mathematics Richard James, University of Minnesota, USA 5:00-5:20 Computational Materials Design: Challenges in Practical Applications Sadasivan Shankar, Harvard University, USA 5:25-5:45 Title Not Available at Time of Publication Michael S. Vogelius, Rutgers University, New Brunswick, USA 5:50-6:10 Title Not Available at Time of Publication Mary Galvin-Donoghue, National Science Foundation, USA
52
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS67
MS68
Volume Penalty and Fourier Methods for PDEs on Irregular Domains Part II of II 4:35 PM-6:15 PM
The Development of Unstructured High Order Methods for Industrial Scaleresolving Simulations 4:35 PM-6:15 PM
Parallel Programming Models, Algorithms and Frameworks for Extreme Computing - Part III of III 4:35 PM-6:15 PM
Room:251 F
Room:254 A
Room:254 B
For Part 1 see MS41 Volume penalty methods allow one to extend the domain that a partial differential equation is defined on (ie. an irregular geometry) to that of a simple rectangular domain through the addition of a volume penalty, forcing term. The resulting penalized equations then provide a flexible and efficient framework for solving (moving) interface and boundary problems via a solution using I) equispaced grids, ii) Nonconforming FEM, iii) Fourier methods. This minisymposium brings together mathematicians and practitioners to discuss both theoretical aspects such as convergence rates and stability, as well as applications and simulations of physical phenomena.
Through their algorithmic efficiency and guaranteed accuracy, unstructured high-order methods (HOM) seem to provide enabling technology for accurate, adaptive and timely scale-resolving simulations in complex geometries. This minisymposium addresses the following topics - development or the adaptation of LES subgrid scale models and hybrid/wall models in close interaction with the discretisation; - assessment on academic benchmarks, including on irregular meshes; - development of hp-adaptive strategies; industrial applications; - practical aspects such as large scale parallellisation, non-matching connections, post or co-processing .... Both finite element-like methods (DGM, SDM, FR, RDS, ...) as well as high- order FVM or FDM are considered.
For Part 2 see MS43 Multicore/manycore processors and accelerators are universally available as both collections of homogeneous standard microprocessors and as attached heterogeneous co-processors. Application and library software developers may often effectively use these processors and some general approaches have emerged. It is widely recognized that careful design of software and data structures, with effective memory management, are the most critical issues to obtain scalable optimized performance on those systems. In these minisymposia we discuss current experiences and development of applications, libraries and frameworks using a variety of hardware. Speakers will address performance results and software design.
Saturday, March 14
MS66
Organizer: David Shirokoff New Jersey Institute of Technology, USA
Organizer: Koen Hillewaert
4:35-4:55 New Active Penalty Methods with Applications to Fluid Flow Jean-Christophe Nave, McGill University, Canada
CENAERO, Belgium
Organizer: Kengo Nakajima
Organizer: Andrea D. Beck
University of Tokyo, Japan
5:00-5:20 Penalty Methods for the Hyperbolic System Modeling the Wall-Plasma Interaction in a Tokamak Philippe Angot, Thomas Auphan, and Olivier Guès, Aix-Marseille Université, France 5:25-5:45 A Dispersionless Fourier Method for the Maxwell Equations Using Volume Penalization Ryan Galagusz, McGill University, Canada 5:50-6:10 Fourier based PDE Solution on Complex Domains Mark Lyon, University of New Hampshire, USA
University of Stuttgart, Germany
Organizer: Michael Heroux
4:35-4:55 The Application of High Order Dgm for Resolved and Wall-Modeled Les of Full Scale Turbomachinery Passages Koen Hillewaert and Corentin Carton de Wiart, CENAERO, Belgium; Guillaume Verheylewegen, Université Catholique de Louvain, Belgium; Ariane Frère, CENAERO, Belgium
Sandia National Laboratories, USA
5:00-5:20 The Use of Residual-Based Compact Schemes for Industrial Les Paola Cinnella, ENSAM, ParisTech, France; Cédric Content and Luca Sciacovelli, Arts et Metiers PARISTECH, France 5:25-5:45 Validation of a High-Order Implicit Les Solver for the Simulation of a Low-Reynolds-Number Vertical-Axis Wind Turbine Samuel Kanner and Per-Olof Persson, University of California, Berkeley, USA 5:50-6:10 Applications of the Spectral/ hp Element Method to Complex Flow Geometries David Moxey, Joaquim Peiro, and Spencer Sherwin, Imperial College London, United Kingdom
Organizer: Serge G. Petiton CNRS/LIFL and INRIA, France 4:35-4:55 Intelligent Iterative Methods: the Future of Parallel and Distributed Runtime Tuned Linear Algebra? Serge G. Petiton, CNRS/LIFL and INRIA, France 5:00-5:20 Performance of Algebraic Multigrid Preconditioners for LargeScale Finite Element Simulations Paul Lin and John Shadid, Sandia National Laboratories, USA 5:25-5:45 Divide and Conquer Algorithms for Large Hermitian Eigenvalue Problems Yousef Saad, University of Minnesota, USA; Vasilis Kalantzis, University of Patras, Greece 5:50-6:10 Unit and Conquer Algorithms for Large Eigenvalue Problems Nahid Emad, University of Versailles, France
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Saturday, March 14
MS69
MS70
One-Shot Methods for Optimization with PDEs 4:35 PM-6:15 PM
Computational Methods for Kinetic Equations and Related Models Part III of V 4:35 PM-6:15 PM
Room:254 C One-shot methods for optimization with PDEs aim at achieving simultaneously convergence of the primal PDE state equation, the adjoint state equation and the design equation. The direction and size of the one-shot optimization steps are usually determined by carefully selected design space preconditioners. It turns out, that one-shot methods enable quite often designs for the computational effort of a small, constant multiple of the effort of a simulation, especially if the primal PDE solve is very expensive. The minisymposium presents recent developments in the field of one-shot methods for optimization with PDEs and its potential by showing large scale applications.
Organizer: Andrea Walther Universität Paderborn, Germany
Organizer: Nicolas R. Gauger Technische Universität Kaiserslautern, Germany 4:35-4:55 Augmenting the One-Shot Framework by Additional Constraints Torsten F. Bosse, Argonne National Laboratory, USA; Andreas Griewank, Humboldt University Berlin, Germany 5:00-5:20 Towards Second Order OneShot Methods in the Context of Shape Calculus Volker H. Schulz, University of Trier, Germany 5:25-5:45 Fixed-Point Iterations for Simultaneous One-Shot Optimization of Unsteady Flows Stefanie Günther, Technical University Kaiserslautern, Germany; Nicolas R. Gauger, Technische Universität Kaiserslautern, Germany; Qiqi Wang, Massachusetts Institute of Technology, USA 5:50-6:10 On An Extension of the Augmented Lagrangian Approach for One-Shot Optimization Andrea Walther, Universität Paderborn, Germany; Nicolas R. Gauger, Technische Universität Kaiserslautern, Germany
Room:150 AB For Part 2 see MS45 For Part 4 see MS95 Kinetic descriptions play an important role in a variety of physical, biological, and even social applications. Unfortunately, the large phase space associated with the kinetic description has in the past made simulations impractical in most settings. However, recent advances in computer resources and numerical algorithms are making kinetic models more tractable, and this trend is expected to continue in the future. The purpose of this minisymposium is to report on the continuing progress on numerical analysis and computational science for kinetic equations. It brings together researchers from applied mathematics, computational science, physics, and engineering communities to discuss their work and exchange ideas.
Organizer: Martin Frank RWTH - Aachen University of Technology, Germany
Organizer: Cory Hauck Oak Ridge National Laboratory, USA
Organizer: Ryan G. McClarren Texas A&M University, USA
Organizer: Jingmei Qiu University of Houston, USA
Organizer: Jeffrey A. Willert Los Alamos National Laboratory, USA 4:35-4:55 High Order Asymptotic Preserving Projective Integration Methods Pauline Lafitte, Ecole Centrale Paris, France; Annelies Lejon, Ward Melis, and Giovanni Samaey, Katholieke Universiteit Leuven, Belgium
continued in next column
53
5:00-5:20 Effective High-Order Diffusive Moment Closures with the StaRMAP Software Benjamin Seibold, Temple University, USA; Martin Frank, RWTH - Aachen University of Technology, Germany 5:25-5:45 An Asymptotic-preserving Scheme for Linear Kinetic Equation with Fractional Diffusion Limit Li Wang, University of California, Los Angeles, USA 5:50-6:10 Energy-Conserving Schemes for Vlasov-Type Systems Yingda Cheng, Michigan State University, USA
54
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
MS71 Advances in Large-scale Forward and Inverse Ice Sheet Modeling - Part II of II 4:35 PM-6:15 PM Room:150 DE For Part 1 see MS46 Model-based projections of the dynamics of the polar ice sheets play a central role in anticipating future sea level rise. However, a number of mathematical and computational challenges place significant barriers on improving predictability of these models. These include complex and very highaspect ratio geometries, highly nonlinear and anisotropic rheology, extremely illconditioned (non)linear systems, broad range of length scales, and unknown model parameters that must be inferred from heterogeneous observations, leading to an ill-posed inverse problem and the need to quantify uncertainties in its solution. Speakers in this minisymposium will address these challenges and present recent developments aimed at overcoming them.
Organizer: Noemi Petra University of California, Merced, USA
Organizer: Omar Ghattas University of Texas at Austin, USA
Organizer: Irina Kalashnikova Sandia National Laboratories, USA
Organizer: Georg Stadler Courant Institute of Mathematical Sciences, New York University, USA 4:35-4:55 Multilevel Methods for Forward and Inverse Ice Sheet Modeling Toby Isaac, University of Texas at Austin, USA; Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA; Omar Ghattas, University of Texas at Austin, USA 5:00-5:20 Assessment of Finite Element Schemes for Accurate Modeling of the Grounding Line Mathieu Morlighem, University of California, Irvine, USA; Jerome Monnier, Institut de Mathématiques de Toulouse, France; Helene Seroussi, California Institute of Technology, USA; Nathan Martin, CNRS, France
continued in next column
5:25-5:45 Advances on Ice-Sheet Model Initialization using the First Order Model Mauro Perego, Sandia National Laboratories, USA; Stephen Price, Los Alamos National Laboratory, USA; Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA; Michael S. Eldred, Sandia National Laboratories, USA; Charles Jackson, University of Texas at Austin, USA; John D. Jakeman, Irina Kalashnikova, and Andrew Salinger, Sandia National Laboratories, USA 5:50-6:10 Uncertainty Quantification for Large-Scale Bayesian Inverse Problems with Application to Ice Sheet Models Noemi Petra, University of California, Merced, USA; Toby Isaac, University of Texas at Austin, USA; Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA; Omar Ghattas, University of Texas at Austin, USA
Saturday, March 14
MS72 Career Fair: Alternatives to Academia - Part III of III 4:35 PM-6:15 PM Room:255 For Part 2 see MS47 The career fair will feature representatives from nonacademic employers from industry and government. These representatives will be prepared to discuss with you the opportunities for internships, postdoctoral appointments and full-time jobs at their organizations.
Organizer: William G. Kolata SIAM, USA
Organizer: Kristin O’Neill SIAM, USA This is the most current list at time fo printing. • Argonne National Laboratory • Boeing • Hewlett-Packard Laboratories • Kitware • Lawrence Berkeley National Laboratory • Lawrence Livermore National Laboratory • MathWorks • MIT Lincoln Laboratory • National Institute of Standards & Technology • NSA • NVIDIA • Oak Ridge National Laboratory • Quantlab The most current list of participating companies is available at www.siam. org/meetings/cse15/career.php.
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
MS73 Efficient and Accurate Solution Techniques for Variable Coefficient Elliptic Partial Differential Equations - Part II of II 4:35 PM-6:15 PM Room:260 A For Part 1 see MS48 Efficient and accurate numerical methods for solving variable coefficient partial differential equations are important for many applications such as seismic imaging and metamaterial design. A variety of techniques including finite element, discontinuous Galerkin, and integral equation methods have been developed to tackled these challenging problems. This session brings together researchers from a broad range of research communities in an effort to build an understanding of the different techniques and open problems in the field.
Organizer: Adrianna Gillman Rice University, USA
Organizer: Lise-Marie ImbertGérard Courant Institute of Mathematical Sciences, New York University, USA 4:35-4:55 Generalized Plane Waves Adapted to Varying Coefficients Lise-Marie Imbert-Gérard, Courant Institute of Mathematical Sciences, New York University, USA 5:00-5:20 Fast Multipole Method as a Preconditioner for Finite Discretizations of Elliptic Boundary Value Problems Huda Ibeid, Rio Yokota, and David E. Keyes, King Abdullah University of Science & Technology (KAUST), Saudi Arabia 5:25-5:45 Fast Direct Solver based on the Cyclic Reduction Algorithm and Hierarchical Matrix Arithmetic for the Solution of Variable-coefficient Elliptic PDEs Gustavo Chavez, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; George M. Turkiyyah, American University of Beirut, Lebanon; Rio Yokota and David E. Keyes, King Abdullah University of Science & Technology (KAUST), Saudi Arabia continued in next column
5:50-6:10 A Polarized-Trace Preconditioner for 2D Helmholtz and Frequency Domain Full-Waveform Inversion Russell Hewett, Total E&P, USA; Leonardo Zepeda-Nunez and Laurent Demanet, Massachusetts Institute of Technology, USA
55
Saturday, March 14
MS74 Large-Scale Inversion and Uncertainty Mitigation 4:35 PM-6:15 PM Room:260 B On top of the conventional challenges that inversion introduces, such as appropriate incorporation of a-priori information and experimental design, inversion of largescale problems requires consideration of the interplay between model reduction and model misspecification as well as means for uncertainty quantification and mitigation. In this minisymposium session, we shall explore various algorithmic strategies to handle these aspects.
Organizer: Lior Horesh IBM T.J. Watson Research Center, USA
Organizer: Stephen Becker University of Colorado Boulder, USA 4:35-4:55 Fast Algorithms for Linear Inverse Problems with Gaussian Priors Kenneth L. Ho and Lexing Ying, Stanford University, USA 5:00-5:20 Accelerating MCMC with Parallel Local Approximations Patrick R. Conrad and Youssef M. Marzouk, Massachusetts Institute of Technology, USA; Natesh Pillai, Harvard University, USA; Aaron Smith, University of Ottawa, Canada 5:25-5:45 Random Matrix Models for the Representation of Model Inadequacy: A Case Study in Chemical Kinetics Rebecca Morrison and Robert D. Moser, University of Texas at Austin, USA 5:50-6:10 Convex Relaxations of Polynomial Imaging Problems Laurent Demanet and Augustin Cosse, Massachusetts Institute of Technology, USA
56
Saturday, March 14
MS75
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
MS76
Saturday, March 14
MS77
Advances in Radial Basis Function and Other Meshfree Methods Part III of V 4:35 PM-6:15 PM
Computational Models of Cardiac Growth and Remodeling 4:35 PM-6:15 PM Room:151 DE
Room:151 G
Room:151 AB
Understanding the processes of cardiac growth and remodeling is of substantial clinical relevance, in particular for improved treatment of heart failure. Computational models for soft tissue growth have been established in the framework of continuum mechanics, and are valuable tools for studying growth mechanisms and testing physiological hypotheses. However, model development and analysis is challenging, in particular because of the complexity and extreme multiscale nature of the processes involved. In this minisymposium we address recent developments in computational models of growth and remodeling. Specific topics will include mathematical model development, numerical solution methods, and clinically relevant applications such as heart failure.
This minisymposium will focus on the software development process used to create Hydra-TH, a CFD code for the Consortium for Advanced Simulation of Light Water Reactors (CASL). We delve into the fine level details of cross- platform builds, effective testing, and collaborative development workflows needed to create a scalable, general purpose CFD code. This includes open source tools to help manage these complexities. Additionally, we include In Situ simulation visualization and analysis along with preprocessing tools ensuring that CFD analysts can effectively use Hydra-TH for performing their desired work. An entire workflow including software and model analysis development will be presented.
For Part 2 see MS50 For Part 4 see MS101 Meshfree methods have many inherent properties that make them useful for a variety of problems in science and engineering, from fitting data to numerically solving differential equations. Such methods offer geometric flexibility, non-uniform resolution, and advantageous trade-offs between accuracy and computational costs. This minisymposium focuses primarily on methods based on radial basis functions and other more general kernels. The talks will address recent advances in the application of these methods to large-scale problems in biology, geophysics, image processing, and finance, as well as theoretical advances in the methods themselves.
Organizer: Varun Shankar University of Utah, USA
Organizer: Grady B. Wright Boise State University, USA 4:35-4:55 RBF-Based Partition of Unity Collocation Methods for the Numerical Solution of PDEs Elisabeth Larsson, Uppsala University, Sweden; Alfa Heryudono, University of Massachusetts, Dartmouth, USA 5:00-5:20 Multilevel Collocation with Radial Basis Functions Patricio Farrell, Weierstrass Institute for Applied Analysis and Stochastics, Germany 5:25-5:45 Kernel-based Image Reconstruction from Scattered Radon Data Armin Iske, University of Hamburg, Germany 5:50-6:10 Dealing with Multiple Boundary Conditions in the RBF Collocation Method Alfa Heryudono, University of Massachusetts, Dartmouth, USA
Organizer: Joakim Sundnes
Software Process for a CASL Sustainable Simulation Software Solution 4:35 PM-6:15 PM
Organizer: Bill Hoffman Kitware, Inc., USA
Simula Research Laboratory, Norway
4:35-4:55 Software Quality with the Open Source Tools CMake, CDash, CTest Bill Hoffman, Kitware, Inc., USA
4:35-4:55 Modeling Growth and Remodeling in Heart Muscle Tissue Joakim Sundnes, Simula Research Laboratory, Norway
5:00-5:20 Tribits: Tribal Build, Integrate, and Test System Roscoe Bartlett, Oak Ridge National Laboratory, USA
5:00-5:20 A Computational Model of Reverse Cardiac Growth in Response to Mechanical Stimulus Lik Chuan Lee, Michigan State University, USA
5:25-5:45 Developing Hydra-TH: A Vertical, VERA-integrated Application based on the Hydra Toolkit Mark Christon, Jozsef Bakosi, and Markus Berndt, Los Alamos National Laboratory, USA; Andrew Bauer, Kitware, Inc., USA; Alan Stagg, Oak Ridge National Laboratory, USA; Balasubramanya Nadiga, Los Alamos National Laboratory, USA; Patrick O’Leary, Kitware, Inc., USA
Simula Research Laboratory, Norway
Organizer: Samuel Wall
5:25-5:45 Human Fetal Growth Model of Hypoplastic Left Heart Syndrome Adarsh Krishnamurthy, University of California, San Diego, USA 5:50-6:10 Finite Element Models of Growth and Remodelling in the Infarct Injured Left Ventricle Samuel Wall, Simula Research Laboratory, Norway
5:50-6:10 Computational Model Builder and ParaView Catalyst: Empowering HPC Workflows Andrew Bauer, Patrick O’Leary, Robert O’Bara, and Berk Geveci, Kitware, Inc., USA
2015 SIAM Conference on Computational Science and Engineering
Saturday, March 14
Professional Development Evening Getting Started with Interdisciplinary Research: Challenges and Opportunities
6:30 PM-7:30 PM Room:355 D Chair: Maria Emelianenko, George Mason University, USA Chair: Cammey Cole Manning, Meredith College, USA Chair: Sven Leyffer, Argonne National Laboratory, USA Join us for an evening devoted to developing and sustaining a successful interdisciplinary research program. Several professionals from academia, government, and industry will share their perspective on challenges and rewards of an interdisciplinary research, training opportunities for students and early career researchers, how to identify open problems and find common language with colleagues in a different field and funding opportunities among other topics. The target audience for this event includes early career professionals (i.e., less than five years past last degree), postdocs, and students. However, we are also encouraging participation from the senior professional community during the networking session. This networking session will take place immediately following this panel. Panelists: Richard Braun
57
Saturday, March 14 Professional Development Evening: Networking Reception 7:30 PM-8:30 PM
Saturday, March 14
Room:355 D
Interdisciplinary Research:
Professional Development Evening Sustaining a Successful Program
8:30 PM-9:30 PM Room:355 D Chair: Maria Emelianenko, George Mason University, USA Chair: Cammey Cole Manning, Meredith College, USA Chair: Sven Leyffer, Argonne National Laboratory, USA Join us for an evening devoted to developing and sustaining a successful interdisciplinary research program. Several professionals from academia, government, and industry will share their perspective on challenges and rewards of an interdisciplinary research, training opportunities for students and early career researchers, how to identify open problems and find common language with colleagues in a different field and funding opportunities among other topics. The target audience for this event includes early career professionals (i.e., less than five years past last degree), postdocs, and students. However, we are also encouraging participation from the senior professional community during the networking session. This networking session will take place immediately prior to this panel. Panelists: Andrea Bertozzi
University of Delaware, USA
University of California, Los Angeles, USA
Thomas Grandine
Fariba Fahroo
The Boeing Company, USA C.T. Kelley
Air Force Office of Scientific Research, USA
North Carolina State University, USA
Omar Ghattas
Carol Woodward
Jeffrey Hittinger
Lawrence Livermore National Laboratory, USA
University of Texas at Austin, USA Lawrence Livermore National Laboratory, USA Wil Schilders
Technische Universiteit Eindhoven, The Netherlands
58
Sunday, March 15 Student Days: Student Chapter Meeting with SIAM Leadership (by invitation only) 7:00 AM-8:00 AM Room:255 D
Registration 7:30 AM-5:00 PM Room:East Foyer
Announcements 8:10 AM-8:15 AM Room:355
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
IP3
MT1
Petascale Finite Element Simulation of Real World’s Complex Structure with Billions DOFs Model 8:15 AM-9:00 AM
Minitutorial: Python Visual Analytics for Big Data Part I of II 9:10 AM-10:50 AM
Room:355
For Part 2 see MT2 Chair: Jonathan Woodring, Los Alamos National Laboratory, USA
Chair: Feng Xiao, Tokyo Institute of Technology, Japan Leading supercomputers offer the computing power of petascale, and exascale systems are expected to be available by the end of this decade. Supercomputers with more than tens of thousands of computing nodes, each of which has many cores cause serious problems in practical finite element software. We have been developing an open source parallel finite element software known as ADVENTURE, which enables very precise analyses of practical structures and machines using over 100 million to billions DOFs mesh. The basic parallel solution algorithms employed are the hierarchical domain decomposition method with balancing domain decomposition as preconditioner. In this talk, I explain several key technologies and one practical application, i.e. seismic response of nuclear power plant subjected to a strong earthquake.
Shinobu Yoshimura University of Tokyo, Japan
Intermission 9:00 AM-9:10 AM
Room:355
Python is a powerful development, computational, and programming environment due to the wide variety of libraries developed for it, and importantly, the enthusiastic, active development and user community. One of the areas where Python excels is visualization and analysis of big data, due to several high-quality modules for both simple and advanced visual analytics. This tutorial will cover the following bigdata visualization capabilities in Python: interactive plotting with IPython, matplotlib, and databases, building web visualizations with Bokeh, and Python integration with VTK and ParaView. Additional information will be provided on mapreduce and NoSQL capabilities. This tutorial is intended for intermediate-level participants who have a basic understanding of the Python language and development environment (i.e., the student ought to have an understanding of native (and ideally numpy) data structures, file I/O, and is able to develop and run simple programs). Beginner participants are welcome, but Python fundamentals, such as language constructs, “hello world,” and program execution will not be covered in this tutorial. Topic A: IPython + Matplotlib using SQlite Joseph Cottam, Indiana University, USA Topic B: NumPy + SciPy + Pandas using Postgres & HDF5 Jonathan Woodring, Los Alamos National Laboratory, USA
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS78
MS79
Teaching Computational Thinking and Practice 9:10 AM-10:50 AM
Large Scale Solution Methods for Coupled and Nonlinear Problems 9:10 AM-10:50 AM
Room:355 A As dependence on computational tools increases, so does the need for better computational training. We discuss our own efforts to provide such training. We address several larger questions including: What is the role of computational training in our various fields? How can we scale training to meet demand? What technologies, languages, principles, and practices should we teach? What is the right balance between conceptual learning and hands-on training and practice? More generally, how should we train the next generation of scientists, statisticians, and engineers in computational methods and practices?
Room:355 D
University of California, Berkeley, USA
The devlopment of solution methods for coupled (e.g. fluid structure interaction) and nonlinear (e.g. nonlinear mechanics) problems is highly challenging. In particular in the context of large scale problems, this might require non-standard approaches for discretization and solution methods, as the simple combination of standard methods for the respective sub-problems or the application of, e.g., Newton’s-method, might lead to stability, efficiency, or scalability constraints. In this minisymposium, we therefor focus on solution methods, which are a priori designed for the solution of coupled and non-linear problems on parallel machines.
Organizer: Philip B. Stark
Organizer: Rolf Krause
Organizer: Kenneth J. Millman
University of California, Berkeley, USA
University of Lugano, Switzerland
9:10-9:30 Teaching Statistical Computing to Undergraduates Kenneth J. Millman and Philip B. Stark, University of California, Berkeley, USA
Organizer: Johannes Steiner
9:35-9:55 Opportunities and Experiences with Teaching Computational Science from the Very Start of University Studies Hans Petter Langtangen, Simula Research Laboratory and University of Oslo, Norway 10:00-10:20 Teaching Data Science from a Computer Science Perspective: Experience from a First Mooc Bill Howe, University of Washington, USA 10:25-10:45 Teaching Computing to Engineers Lorena A. Barba, George Washington University, USA
University of Lugano, Switzerland 9:10-9:30 A Scalable Monolithic Solver for the Coupling of a Finite Element and a Finite Volume Method for Fluid-Structure-Interaction Johannes Steiner and Rolf Krause, University of Lugano, Switzerland 9:35-9:55 Parallel Scalable FETI Methods for Nonlinear Problems Axel Klawonn and Martin Lanser, Universität zu Köln, Germany; Oliver Rheinbach, Technische Universitaet Bergakademie Freiberg, Germany 10:00-10:20 Adaptive Spectral Deferred Correction Methods for Cardiac Simulation Martin Weiser, Zuse Institute Berlin, Germany
continued in next column
59
10:25-10:45 Fluid-Structure-Interaction in Computational Hemodynamcis using Nonlinear Hyperelastic Arterial Wall Models Daniel Balzani, University of DuisburgEssen, Germany; Simone Deparis, École Polytechnique Fédérale de Lausanne, Switzerland; Simon Fausten, University of Duisburg-Essen, Germany; Davide Forti, École Polytechnique Fédérale de Lausanne, Switzerland; Alexander Heinlein and Axel Klawonn, Universität zu Köln, Germany; Alfio Quarteroni, École Polytechnique Fédérale de Lausanne, Switzerland; Oliver Rheinbach, Technische Universitaet Bergakademie Freiberg, Germany; Jörg Schröder, University of Duisburg-Essen, Germany
60
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS81
MS82
High-dimensional Approximation and Integration: Analysis and Computation - Part I of V 9:10 AM-10:50 AM
Resilience in Numerical Simulations and Algorithms at Extreme Scale: Part I of IV - General Topics 9:10 AM-10:50 AM
Room:250 B
Room:250 C For Part 2 see MS107 The advent of extreme scale machines will require the use of parallel resources at an unprecedented scale, probably leading to a high rate of hardware faults. Handling fully these faults at the computer system level may have a prohibitive cost. High performance computing applications that aim at exploiting all these resources will thus need to be resilient, in this minisympoisum different complementary approaches and methods will be presented to possibly address this key aspect of extreme scale computing.
Technische Universität München, Germany
For Part 2 see MS106 This minisymposium focuses on the fundamental problem of how to accurately approximate solutions of both forward and inverse high-dimensional functions. Predicting the behavior of complex phenomena relies on constructing solutions in terms of high dimensional spaces, particularly in the case when the input data (coefficients, forcing terms, initial and boundary conditions, geometry) are affected by large amounts of uncertainty. The resulting explosion in computational effort is a symptom of the curse of dimensionality and this symposium aims at exploring breakthroughs in sparse sampling and representations, nonlinear and greedy approximations, compressed sensing, multilevel methods, and “best-N-term” approximations.
Organizer: Ulrich J. Ruede
Organizer: Clayton G. Webster
University of Erlangen-Nuremberg, Germany
Oak Ridge National Laboratory, USA
Speakers To Be Announced
Oak Ridge National Laboratory, USA
Sunday, March 15
MS80 Computational Engineering (BGCE) Student Paper Prize - Part I of II 9:10 AM-10:50 AM Room:250 A For Part 2 see MS105 The 5th Bavarian Graduate School in Computational Engineering (BGCE) Student Paper Prize will be awarded at the 2015 SIAM CS&E Conference for outstanding student work in the field of Computational Science and Engineering. Eligible for the prize will be undergraduate and graduate students prior to receiving their PhD. Candidates are required to summarize their work in a short paper of at most 4 pages. The prize finalists will present their work in this minisymposium. The prize award announcement will be scheduled at one of the last days of the conference.
Organizer: Tobias Neckel Technische Universität München, Germany
Organizer: Hans-Joachim Bungartz
Organizer: Guannan Zhang Organizer: Max Gunzburger Florida State University, USA
Organizer: Albert Cohen Université Pierre et Marie Curie, France 9:10-9:30 A Theory for Model Verification Ronald DeVore, Texas A&M University, USA 9:35-9:55 High-Order Digital Nets for Parametric and Stochastic Operator Equations Christoph Schwab, ETH Zürich, Switzerland 10:00-10:20 Low-Rank Adaptive Tensor Approximation Wolfgang Dahmen, RWTH Aachen, Germany 10:25-10:45 Multivariate Decomposition Methods $\infty$Variate Problems Grzegorz W. Wasilkowski, University of Kentucky, USA
Organizer: Keita Teranishi Sandia National Laboratories, USA
Organizer: Luc Giraud INRIA, France
Organizer: Michael Heroux Sandia National Laboratories, USA
Organizer: Emmanuel Agullo INRIA, France 9:10-9:30 Resilient Programming Models Michael Heroux, Sandia National Laboratories, USA 9:35-9:55 Title Not Available at Time of Publication George Bosilca, University of Tennessee, Knoxville, USA 10:00-10:20 Portable Programming and Runtime Support for ApplicationControlled Resilience in Large-Scale Scientific Applications Andrew A. Chien, University of Chicago and Argonne National Laboratory, USA; Hajime Fujita, Zachary Rubenstein, Nan Dun, Aiman Fang, and Ziming Zheng, University of Chicago, USA 10:25-10:45 MPI Fault Tolerance: The Good, The Bad, The Ugly Martin Schulz, Lawrence Livermore National Laboratory, USA
2015 SIAM Conference on Computational Science and Engineering
61
Sunday, March 15
Sunday, March 15
Sunday, March 15
MS83
MS84
MS85
Preconditioning and Iterative Methods for Linear Systems - Part I of II 9:10 AM-10:50 AM
Error Analysis and Scalability of UQ Methodologies for Inverse Problems - Part I of II 9:10 AM-10:50 AM
Advanced Finite Element Methods for Nonlinear Materials and Fluids Part I of III 9:10 AM-10:50 AM
Room:250 E
Room:250 F
For Part 2 see MS109 The application of novel uncertainty quantification (UQ) methodologies to solve stochastic inverse problems for a physical system has improved the predictive capabilities of many computationally complex models. However, a full error analysis and study of scalability for the combined UQ method and computational model is often lacking. This session addresses the issues of error analysis and scalability of several UQ methods applied to stochastic inverse problems associated with specific complex physical systems. Both a priori and a posteriori error analyses of deterministic and stochastic sources of error are considered along with computational challenges in scaling such analyses.
For Part 2 see MS110 Numerical modeling of materials and fluids with nonlinear behavior plays an increasing role in computational science and engineering, particularly in the highfidelity simulation of elastic and plastic deformations, magneto-sensitive materials, liquid crystals, biological tissues and blood flow, among others. Despite a long history of efforts, there is still a need for the development of advanced finite-element methods with specific properties such as accurate representation of stresses, forces, and other quantities depending on the application. In many of these applications, mesh-adaptive implementations and efficient solvers are an important issue due to the size and complexity of the problems.
Organizer: Troy Butler
Organizer: James H. Adler
University of Colorado, Denver, USA
Tufts University, USA
Organizer: Steven Mattis
Organizer: Gerhard Starke
University of Texas at Austin, USA
University of Duisburg-Essen, Germany
9:10-9:30 Quantifying Errors in a Probabilistic Solution to Stochastic Inverse Problems for Physics-Based Models Troy Butler, University of Colorado, Denver, USA
Organizer: Thomas Manteuffel
Room:250 D For Part 2 see MS108 The solution of large, sparse linear systems remains the limiting component of many problems. Iterative solvers are often the methods of choice for such systems, and effective preconditioning is usually necessary to achieve an acceptable speed of convergence. In this minisymposium we focus on recent developments in preconditioning, the convergence of solvers and the application of iterative methods.
Organizer: Jennifer Pestana University of Manchester, United Kingdom
Organizer: Andy Wathen University of Oxford, United Kingdom 9:10-9:30 Convergence of Stationary Iteration with Indefinite Preconditioner Andy Wathen, University of Oxford, United Kingdom 9:35-9:55 Preconditioning for Various Cahn-Hilliard Systems Jessica Bosch and Martin Stoll, Max Planck Institute, Magdeburg, Germany 10:00-10:20 Null-Space Based Preconditioners for Saddle-Point Systems Ron Estrin and Chen Greif, University of British Columbia, Canada 10:25-10:45 Unreduced Symmetric KKT Systems Arising from Interior Point Methods Mattia Tani and Valeria Simoncini, Universita’ di Bologna, Italy; Benedetta Morini, Universita’ di Firenze, Italy
9:35-9:55 Adaptive MeasureTheoretic Inverse Techniques for High Dimensional Parameter Domains and Complex Multi-Scale Models Lindley C. Graham, University of Texas at Austin, USA; Troy Butler, University of Colorado, Denver, USA; Clint Dawson, University of Texas at Austin, USA 10:00-10:20 Optimizing Quantities of Interest in High Dimensions to Improve Solutions to Inverse Problems Scott Walsh, University of Colorado, Denver, USA 10:25-10:45 Region of Influence Sensitivty Analysis for Time Dependant Problems Varis Carey and Robert D. Moser, University of Texas at Austin, USA
University of Colorado Boulder, USA 9:10-9:30 First-Order System Least Squares for Isotropic and Anisotropic Materials in Hyperelasticity Benjamin Müller, Gerhard Starke, and Jörg Schröder, University of Duisburg-Essen, Germany 9:35-9:55 Momentum Balance Accuracy in Finite Element Methods for Elastoplasticity Gerhard Starke, University of DuisburgEssen, Germany 10:00-10:20 Modeling MagnetoMechanical Interactions in Deformable Solids James H. Adler, Luis Dorfmann, Dong Han, Scott Maclachlan, and Chris Paetsch, Tufts University, USA 10:25-10:45 Advanced Finite Element Methods for ChemoElectromechanical Skeletal Muscle Mechanics Oliver Rohrle, Thomas Heidlauf, Mylena Mordhorst, and Daniel Wirtz, University of Stuttgart, Germany
62
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
MS86 Adjoint Methods for Unsteady and Chaotic PDEs and Large-scale Optimization and Control Part I of II 9:10 AM-10:50 AM Room:251 A For Part 2 see MS111 We expose a growing intersection of optimization and control using largescale, unsteady PDE-based simulations. This intersection has traditionally been small because most practitioners perform optimization and control using inexpensive, reduced order models; however, this is changing. Current-day simulations solving unsteady and chaotic PDEs, such as the Navier-Stokes equations, are leveraging advance in high performance computing. Petascale, and soon exascale, supercomputers are increasingly available to researchers, enabling them to attempt optimization and control directly using large-scale simulations. The resulting area of large-scale optimization and control, particularly with the aid of the unsteady adjoint method, is the theme of this proposed minisymposium.
Organizer: Qiqi Wang Massachusetts Institute of Technology, USA
Organizer: Nicolas R. Gauger Technische Universität Kaiserslautern, Germany
Organizer: Daniel J. Bodony University of Illinois at Urbana-Champaign, USA 9:10-9:30 Optimal Wall-Forcing for Compressible Wall-Bounded Flows Using Adjoint Techniques Taraneh Sayadi and Peter Schmid, Imperial College London, United Kingdom 9:35-9:55 A Coupled CFD – CAA Adjoint Method for Aeroacoustic Optimization and Error Estimation Enrico Fabiano, Dimitri Mavriplis, and Jay Sitaraman, University of Wyoming, USA
continued in next column
10:00-10:20 Efficient Approaches for Optimal Active Flow Control Nicolas R. Gauger, Technische Universität Kaiserslautern, Germany; Anil Nemili, Emre Özkaya, and Stefanie Günther, Technical University Kaiserslautern, Germany 10:25-10:45 Aerodynamic Design for Unsteady Flows Using An Adjoint Approach Eric Nielsen, NASA Langley Research Center, USA; Boris Diskin, National Institute of Aerospace, USA
Sunday, March 15
MS87 Theory Implementation and Applications of HDG Methods - Part II of II 9:10 AM-10:50 AM Room:251 B For Part 1 see MS62 In this minisymposium, we will discuss the latest advancements related to the hybrid discontinuous Galerkin (HDG) method. HDG applies a static condensation technique within the DG framework, so that the only globally coupled degrees of freedom are those located on the mesh skeleton or trace space, greatly reducing the global system size. The method is also promising for its applicability to current and emerging parallel architectures. The HDG method has proven to be a popular method and has, in recent years, been applied in the context of steady-state diffusion, Maxwell’s equations, convection-diffusion problems, elastodynamics, Stokes and incompressible Navier-Stokes equations.
Organizer: Hari Sundar University of Utah, USA
Organizer: Tan Bui-Thanh University of Texas at Austin, USA
Organizer: Cuong Nguyen Massachusetts Institute of Technology, USA 9:10-9:30 Multiscale Hybridizable Discontinuous Galerkin Methods Cuong Nguyen, Joel Saa-Seoane, David Moro, Francisco J. Roca, and Jaime Peraire, Massachusetts Institute of Technology, USA 9:35-9:55 To CG or HDG: Updates on Our Comparative Study Mike Kirby and Sergey B. Yakovlev, University of Utah, USA 10:00-10:20 Stable and Robust Hybridized Discontinuous Galerkin Methods for High Reynolds Number Flow Problems Krzysztof Fidkowski and Johann Dahm, University of Michigan, USA 10:25-10:45 A Computational Framework for Target-Based HP-Adaptation in Compressible Flow Simulation Using HDG Methods Georg May and Michael Woopen, RWTH Aachen University, Germany
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS88
MS89
Computational Advances in Energy Research
AWM Meeting - Workshop: Research Talks by Recent Ph.D.s:Mathematical Modeling and HighPerformance Computing for Multiscale and Multiphysics Problems - Part I of II 9:10 AM-10:50 AM
9:10 AM-10:50 AM Room:251 C Part of the SIAM Workshop Celebrating Diversity Computational modeling for geoscience applications has never been more relevant or fundamental for a range of problems in energy extraction and recovery. This session highlights advances in modeling for recent and more traditional energy applications. Unconventional resource modeling work discussed in this session includes microseismic event location in the context of hydraulic fracturing and modeling of methane hydrates. New advances in traditional seismic wave modeling include more efficient schemes for implementation of absorbing boundary conditions and comparison of finite difference solutions of the wave equation to solutions from discontinuous Galerkin finite element methods.
Organizer: Susan E. Minkoff University of Texas at Dallas, USA
Organizer: Raegan Higgins Texas Tech University, USA
Organizer: Stephen Wirkus Arizona State University, USA 9:10-9:30 Microseismic Event Location Via Full Waveform Inversion Susan E. Minkoff, University of Texas at Dallas, USA 9:35-9:55 Comparison Between DG and Finite Difference Methods for Acoustics with Smooth Coefficients Mario Bencomo, Rice University, USA 10:00-10:20 Partitioned Low Rank fast and Efficient Compression of Absorbing Boundary Conditions for the Helmholtz Equation Rosalie Belanger-Rioux and Laurent Demanet, Massachusetts Institute of Technology, USA 10:25-10:45 Analysis and Numerical Approximation for Adsorption Models Francis P. Medina and Malgorzata Peszynska, Oregon State University, USA
Room:251 D For Part 2 see MS114 Sponsored by Association for Women in Mathematics (AWM) This minisymposium highlights selected research talks by female recent Ph.D.s in developing multiscale multiphysics models, computational methodologies, and highperformance computing. Advances in science and engineering—-in disciplines ranging from health to energy and the environment to defense—-rely on predictive measurement and analysis of multiscale multiphysics systems. Such advances are essential for scientific discovery, engineering design and policy making. The participants of the minisymposium will discuss advanced computational algorithms and simulation strategies for solving modern and nextgeneration systems that require analyzing complex physical phenomena such as fluidsolid interactions, mechanics of biological materials, moleculer motions, charge transport, ocean and cloud dynamics, and oil recovery process. The topics include physical modeling, mathematical analysis, numerical analysis, algorithms, implementation, performance, and scalability. This minisymposium is part of the Association for Women in Mathematics(AWM) workshop that will promote cross-fertilization of ideas among women scientists in multiple disciplines.
Organizer: MiSun Min Argonne National Laboratory, USA
Organizer: Fengyan Li Rensselaer Polytechnic Institute, USA
Organizer: Jingmei Qiu University of Houston, USA
continued in next column
63
9:10-9:30 Recovering Exponential Accuracy in Spectral Methods Involving Piecewise Smooth Functions with Unbounded Derivative Singularities Zheng Chen, Iowa State University, USA 9:35-9:55 Efficient High-Order Algorithms for Solving Drift-Diffusion Systems Ying He, University of California, Davis, USA 10:00-10:20 Estimating Residual Stresses in Arteries by an Inverse Spectral Technique Sunnie Joshi, Temple University, USA 10:25-10:45 Force-based Blended Atomistic-to-continuum Coupling Method for Crystals: Theory and Computations Xingjie Li, Brown University, USA
64
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS90
MS91
MS92
Software and Methods for Spatial Modeling and Simulation in Systems Biology - Part I of II 9:10 AM-10:50 AM
Parametric Model Reduction and Inverse Problems - Part I of IV 9:10 AM-10:50 AM Room:251 F
Applied Mathematics and Computation in the Pharmaceutical Industry Part I of II 9:10 AM-10:50 AM
Room:251 E
For Part 2 see MS116 In recent years model reduction techniques have been used to efficiently solve forward problems given, for instance, as PDEs, dynamical systems or complex networks. In this minisymposium we want explore applications of model reduction in the context of inverse problems.
Sunday, March 15
Room:254 A
Virginia Tech, USA
For Part 2 see MS117 The pharmaceutical industry continues to rapidly develop analytical capabilities tackling scientific, engineering, manufacturing and market challenges. We highlight current mathematical and computational frontiers and areas of interest to the industry in this minisymposium. Specific areas include modeling and simulation, optimization, image processing and data mining/modeling. Modeling and simulation is used in both scientific and business applications within the industry. We highlight disease and treatment models as well as models underlying objective functions in optimization. Data mining/ modeling is now driven by large amounts of biomarker data varying from next generation sequencing to ultra-high resolution imaging.
Organizer: Andreas Hellander
Organizer: Serkan Gugercin
Organizer: Jeffrey Saltzman
Uppsala University, Sweden
Virginia Tech, USA
AstraZeneca, USA
Organizer: Brian Drawert
9:10-9:30 Multiscale Model Reduction for PDE-Constrained Optimization Eldad Haber, University of British Columbia, Canada; Lars Ruthotto, Emory University, USA
9:10-9:30 Numerical Solutions of a Partial Differential Equations in a Pharmacometric Context Jeffrey Saltzman, AstraZeneca, USA
For Part 2 see MS115 Simulation of spatially inhomogeneous biological systems is a challenging problem that has motivated the development of various software packages. Simulation challenges include the multiphysics/multiscale nature of the systems and the need to capture stochastic effects. Modeling challenges include specification of the reaction systems, geometry and mesh generation for complex and dynamic shapes. Software issues include testing and verification, user community issues, and dissemination and promotion. Also, there are the issues of model sharing, validation and reproducibility of results. This minisymposium brings together representatives of different software packages from the computational systems biology community to discuss these challenges.
University of California, Santa Barbara, USA 9:10-9:30 On-Lattice and off-Lattice Hybrid Simulation Using the Smoldyn Software Steven Andrews, Fred Hutchinson Cancer Research Center, USA 9:35-9:55 From Macroscopic to Microscopic Simulations Using Mesord David Fange, Uppsala University, Sweden 10:00-10:20 MCell/CellBlender: An Environment for Spatially Realistic Simulation of Cellular Microphysiology Thomas M. Bartol, The Salk Institute, USA 10:25-10:45 E-Cell System Version 4.0: an Integrated Platform for Singleparticle-level Simulations Kazunari Kaizu, Kozo Nishida, Masaki Watabe, Arjunan Satya, Kazunari Iwamoto, and Koichi Takahashi, RIKEN, Japan
Organizer: Eldad Haber University of British Columbia, Canada
Organizer: Lars Ruthotto Emory University, USA
Organizer: Eric De Sturler Virginia Tech, USA
Organizer: Misha E. Kilmer Tufts University, USA
Organizer: Christopher A. Beattie
9:35-9:55 Model Mis-Specification and Model Reduction - Connecting the Dots Lior Horesh, IBM T.J. Watson Research Center, USA 10:00-10:20 Model Reduction for Some Inverse Problems in Finance Ekkehard W. Sachs, University of Trier, Germany and Virginia Tech, USA; Marina Schneider, University of Trier, Germany 10:25-10:45 Inference for Prediction in Nonlinear Systems Harriet Li, Massachusetts Institute of Technology, USA
9:35-9:55 Applications of Modeling and Simulation in Drug Discovery and Development Chandni Valiathan, Merck & Co., Inc., USA 10:00-10:20 Mathematical Modeling at Two Opposite Ends of the Scale Spectrum with the Same Objective in Mind: Improve Human Health Antonio Cabal, Merck Research Laboratories, USA 10:25-10:45 A Simultaneous Approach to Parameter Estimation with Ode Models: a Case Study with Viral Dynamics Models Khamir Mehta and Junghoon Lee, Merck & Co., Inc., USA
2015 SIAM Conference on Computational Science and Engineering
65
Sunday, March 15
Sunday, March 15
Sunday, March 15
MS93
MS94
MS95
Computational Applications Performing at Petascale Level and Beyond Part I of II 9:10 AM-10:50 AM
Student Days: SIAM Student Chapter Presentations Part I of II 9:10 AM-10:50 AM
Computational Methods for Kinetic Equations and Related Models Part IV of V 9:10 AM-10:50 AM
Room:254 B
For Part 2 see MS119 SIAM Student Chapter presentations.
For Part 2 see MS118 Large-scale simulations are required to validate and then exercise complex multiscale models such as weather forecasting, computational astronomy, or industrial burners. This minisymposium highlights the impact of petascale computing on scientific discoveries by presenting the recent work of several research teams that develop and employ applications to tackle such challenging problems using full capabilities of modern leadership-class supercomputers. Discussions emphasize both the scientific impact of the related work and the challenges associated with computing at such scale.
Organizer: Bilel Hadri King Abdullah University of Science & Technology (KAUST), Saudi Arabia
Organizer: Hatem Ltaief King Abdullah University of Science & Technology (KAUST), Saudi Arabia 9:10-9:30 Exascale: the Why and the How David E. Keyes, King Abdullah University of Science & Technology (KAUST), Saudi Arabia 9:35-9:55 From Optimal Algorithms to Fast Petascale Solvers Björn Gmeiner, University of ErlangenNuremberg, Germany; Holger Stengel, Erlangen Regional Computing Center, Germany; Christian Waluga, Technical University of Munich, Germany; Barbara Wohlmuth, Technische Universität München, Germany; Ulrich J. Ruede, University of Erlangen-Nuremberg, Germany 10:00-10:20 GPU for Adaptive Optics on Ground Based Astronomical Telescopes: Simulations and Real-Time Control Damien Gratadour and Eric Gendron, Observatoire de Paris, Meudon, France; Hatem Ltaief and Hatem Ltaief, King Abdullah University of Science & Technology (KAUST), Saudi Arabia 10:25-10:45 Extreme Scale Solution of Engineering Applications Using Uintah Martin Berzins, University of Utah, USA
Room:254 C
Organizer: Rachel Levy Harvey Mudd College, USA 9:10-9:21 System Architecture for a Cooperative Fleet of Autonomous Underwater Vehicles (AUVs) Stacey Joseph-Ellison, Qi Zhou, Zhaoyang Fu, Zakaria Daud, and Hong Liu, EmbryRiddle Aeronautical University, USA 9:26-9:37 Affirmative Actions in Education Creating Division and Inefficiency Among Beneficiary Groups in India Aprant Ajay, Delhi Technological University, India 9:42-9:53 Interfacial Motion by Mean Curvature in Liquid Crystals Amy Spicer and Apala Majumdar, University of Bath, United Kingdom 9:58-10:09 Theory and Computation for Bilinear Quadratures Christopher Wong, University of California, Berkeley, USA 10:14-10:25 Survival Probability of Beneficial Mutations in Bacteria Anna Zhu and Lindi M. Wahl, Western University, Canada 10:30-10:41 On Strategic Defense in Stochastic Networks Ryan White and J. H. Dshalalow, Florida Institute of Technology, USA
Room:150 AB For Part 3 see MS70 For Part 5 see MS120 Kinetic descriptions play an important role in a variety of physical, biological, and even social applications. Unfortunately, the large phase space associated with the kinetic description has in the past made simulations impractical in most settings. However, recent advances in computer resources and numerical algorithms are making kinetic models more tractable, and this trend is expected to continue in the future. The purpose of this minisymposium is to report on the continuing progress on numerical analysis and computational science for kinetic equations. It brings together researchers from applied mathematics, computational science, physics, and engineering communities to discuss their work and exchange ideas.
Organizer: Martin Frank RWTH - Aachen University of Technology, Germany
Organizer: Cory Hauck Oak Ridge National Laboratory, USA
Organizer: Ryan G. McClarren Texas A&M University, USA
Organizer: Jingmei Qiu University of Houston, USA
Organizer: Jeffrey A. Willert Los Alamos National Laboratory, USA 9:10-9:30 Asymptotic-Preserving Scheme for the Fokker-PlanckMaxwell System in the Quasi-Neutral Regime Stephane Brull, Institut Polytechnique de Bordeaux, France
continued on next page
66
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS95
MS96
MS97
Computational Methods for Kinetic Equations and Related Models Part IV of V 9:10 AM-10:50 AM
Computational Approaches and Multi-scale Modeling of Complex Fluids - Part I of II 9:10 AM-10:50 AM Room:150 DE
Frameworks, Algorithms, and Scalable Technologies for Parallel Unstructured Mesh Workflows - Part I of II 9:10 AM-10:50 AM
continued
For Part 2 see MS122 Complex fluids, which are suspensions of particles in a viscous liquid, can exhibit both viscous and elastic response to external forcing. Capturing the exact interplays between the particles and the surrounding fluid in a continuum modeling presents significant challenges as hydrodynamics interactions happen on short and long scales. Added difficulties arise when such suspensions interact with boundaries, where peculiar behavior has been experimentally observed, or when the surrounding fluid itself is viscoelastic, as it is the case of most biological liquid. This minisymposium’s goal is to explore recent advances, especially on the computational front, on both topics.
Sunday, March 15
9:35-9:55 Solving Kinetic Equations to Model the Core-Collapse Supernova Explosion Mechanism Eirik Endeve and Cory Hauck, Oak Ridge National Laboratory, USA; Yulong Xing, University of Tennessee and Oak Ridge National Laboratory, USA; Tony Mezzacappa, Oak Ridge National Laboratory, USA 10:00-10:20 Versions of Discontinuous Galerkin Algorithms for Diffusion and for Energy-Conserving Hamiltonian Dynamics Ammar Hakim, Princeton Plasma Physics Laboratory, USA; Greg Hammett, Eric Shi, Ian Abel, and Tim Stoltzfus-Dueck, Princeton University, USA 10:25-10:45 Numerical Simulation of the Crookes Radiometer Guillaume Dechristé, Institut de Mathématiques de Bordeaux, France; Luc Mieussens, Universite de Bordeaux I, France
Organizer: Christel Hohenegger University of Utah, USA
Organizer: Enkeleida Lushi Brown University, USA
Room:258 For Part 2 see MS121 This minisymposium presents recent advances in the development of parallel unstructured mesh methods and highlights their use in applications. These technologies are designed and implemented to support reliable simulation technologies that run on today’s leadership-class parallel systems with specific attention to next generation and emerging architectures that will bridge the peta-to-exascale gap. Unstructured meshing technologies presented in these sessions will emphasize memory-efficiency through array based data structures and adaptivity, increased fidelity through high-order methods, and parallelism in mesh generation, adaptation, and in-memory workflows. Many of the minisymposium presentations are associated with the DOE SciDAC FASTMath institute (http://www.fastmathscidac.org/).
9:10-9:30 Stabilizing the Collective Motion of Micro-swimmers using Confinement Enkeleida Lushi, Brown University, USA
Organizer: Cameron Smith
9:35-9:55 Flagellar Activity Influences Self-Organization in Confined Microswimmer Suspensions Alan Cheng Hou Tsang and Eva Kanso, University of Southern California, USA
Rensselaer Polytechnic Institute, USA
10:00-10:20 Swimming Dynamics of Microorganisms in Viscoelastic Fluids Near a Wall Gaojin Li and Arezoo Ardekani, Purdue University, USA 10:25-10:45 Dynamics of MicroSwimmers Inside a Peristaltic Pump Adam Stinchcombe, University of Michigan, USA; Enkeleida Lushi, Brown University, USA
Rensselaer Polytechnic Institute, USA
Organizer: Mark S. Shephard 9:10-9:30 An Array-Based Mesh Topological Representation That Effectively Supports General Mesh Modification Dan A. Ibanez and Mark S. Shephard, Rensselaer Polytechnic Institute, USA 9:35-9:55 Efficient Unstructured Mesh Traversal Methods Based on ArrayBased Half Facets Navamita Ray, Argonne National Laboratory, USA; Xinglin Zhao, Stony Brook University, USA; Vijay Mahadevan, Argonne National Laboratory, USA; Xiangmin Jiao, State University of New York, Stony Brook, USA
continued on next page
2015 SIAM Conference on Computational Science and Engineering
10:00-10:20 Threading Mesh Optimization Codes Using Transactional Memory Barna Bihari and Lori A. Diachin, Lawrence Livermore National Laboratory, USA; Patrick M. Knupp, Sandia National Laboratories, USA 10:25-10:45 M3D-C1 Adaptive Loop Going from 2D Axisymmetric to Full 3D E. Seegyoung Seol, Mark S. Shephard, and Fan Zhang, Rensselaer Polytechnic Institute, USA
67
Sunday, March 15
Sunday, March 15
MS98
MS99
High-Order Methods for Computational Fluid Dynamics - Part I of IV 9:10 AM-10:50 AM
Modeling Microswimmer Locomotion using the Method of Regularized Stokeslets 9:10 AM-10:50 AM
Room:259
Room:260 A
For Part 2 see MS123 Novel contributions in the field of highorder numerical methods for computational fluid dynamics. Presentations will explore new high-order methods, benchmarking of existing schemes, and applications to turbulent flow problems.
Microscopic swimmers live in highly viscous environments where inertial effects are negligible. The way that these organisms have adapted to such unique conditions is of interest to biologists and engineers alike. For instance, a better understanding of microswimmer locomotion may enable the development of novel biomedical technologies, such as mobile biosensors or drug delivery devices. This minisymposium presents numerical studies of microswimmer motility performed using the method of regularized Stokeslets, a method designed for simulating low Reynolds number fluid- structure interactions. Easy to implement, efficient, and highly adaptable, the method of regularized Stokeslets is a powerful tool with broad applicability.
Organizer: Brian C. Vermeire Imperial College London, United Kingdom
Organizer: Antony Jameson Stanford University, USA
Organizer: Peter E. Vincent Imperial College London, United Kingdom 9:10-9:30 Theoretical Aspects of HighOrder Flux Reconstruction Schemes Peter E. Vincent, Imperial College London, United Kingdom 9:35-9:55 Comparison of Continuous, Discontinuous and Hybrid Finite Element Methods for Accuracy and Efficiency Steven R. Allmaras, Marshall Galbraith, and David l. Darmofal, Massachusetts Institute of Technology, USA 10:00-10:20 Adaptive Multiresolution Schemes for Discontinuous Methods Siva Nadarajah and Philip Zwanenburg, McGill University, Canada 10:25-10:45 Goal-Oriented Curved Mesh Optimization for High-Order Finite-Element Methods Krzysztof Fidkowski and Devina Sanjaya, University of Michigan, USA
Organizer: Jonathan H. Tu University of California, Berkeley, USA 9:10-9:30 The Method of Regularized Stokeslets: Motivation and Applications Ricardo Cortez, Tulane University, USA 9:35-9:55 The Effects of Rotation and Translation on Flagellar Synchronization Jonathan H. Tu, Murat Arcak, and Michel M. Maharbiz, University of California, Berkeley, USA 10:00-10:20 Modeling Sperm Motility Using a Kirchhoff Rod Model Sarah D. Olson, Worcester Polytechnic Institute, USA 10:25-10:45 Bacteria Association with Ciliated Surfaces Eva Kanso, University of Southern California, USA
68
Sunday, March 15
MS100 Analysis of Noisy Networks in Theory and Practice Part I of II 9:10 AM-10:50 AM Room:260 B For Part 2 see MS125 Network analysis is an elegant, graph-based method for studying complex systems of interacting entities. However, real-world observed systems inherently include some discrepancy from reality. For example, measurements may not be exact, as not all relations are observed. Since traditional graph theory deals with exact mathematics, one of the most challenging aspects of network analysis is accounting for this noise. This minisymposium will include talks from application experts on how they handle noise, and presentations about the effect of noise in networks generally. We conclude with a discussion on how the general theory and application-specific methods can benefit from each other.
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS101
MS102
Advances in Radial Basis Function and Other Meshfree Methods Part IV of V 9:10 AM-10:50 AM
Parallel, Adaptive Methods for Logically Cartesian Meshes - Part I of II 9:10 AM-10:50 AM
Room:151 AB
For Part 2 see MS127 Adaptive refined Cartesian mesh methods, which started with the seminal paper by Berger-Oliger in 1984 have now been in development for just over 30 years. However, there are still many questions about the scalability of these methods. The meta-data needed to store sub-grid locations can become prohibitive, and dynamic refinement an present a challenging parallel problem. However recent developments in quad tree adaptation and other approaches for Berger-Oliger refinement may overcome some of these problems. In this minisymposium, speakers have been asked to address the scalability of their adaptive frameworks, and share experiences using available packages, including Chombo, Paramesh, p4est, AMRClaw, Overture, GeoClaw and ForestClaw.
Organizer: Sanjukta Bhowmick
For Part 3 see MS75 For Part 5 see MS126 Meshfree methods have many inherent properties that make them useful for a variety of problems in science and engineering, from fitting data to numerically solving differential equations. Such methods offer geometric flexibility, non-uniform resolution, and advantageous trade-offs between accuracy and computational costs. This minisymposium focuses primarily on methods based on radial basis functions and other more general kernels. The talks will address recent advances in the application of these methods to large-scale problems in biology, geophysics, image processing, and finance, as well as theoretical advances in the methods themselves.
University of Nebraska, Omaha, USA
Organizer: Varun Shankar
Organizer: Benjamin A. Miller
University of Utah, USA
Massachusetts Institute of Technology, USA
Organizer: Grady B. Wright
9:10-9:30 Spectral Subgraph Detection in Noisy, Uncertain Networks Benjamin A. Miller, Massachusetts Institute of Technology, USA
Boise State University, USA
9:35-9:55 Epidemic in Time and Space: Modeling Spatial Outbreak Dynamics Armin Mikler, University of North Texas, USA 10:00-10:20 Analysis and Control of Cascading Failures of Power Transmission Systems: a Macro View Daniel Bienstock and Guy Grebla, Columbia University, USA 10:25-10:45 Visual Analytics for Detection and Tracking of Emergent Subgraphs in Social Networks Nadya Bliss and Ross Maciejewski, Arizona State University, USA
9:10-9:30 Oversampling Near the Boundary and Improved Exponential Convergence Rates Barbara Zwicknagl, University of Bonn, Germany 9:35-9:55 Beyond Quasi-unifomity: Kernel Approximation with a Local Mesh Ratio Thomas C. Hangelbroek, University of Hawaii, Manoa, USA 10:00-10:20 Fast Computation of Orthonormal Bases for RBF Native Spaces Stefano De Marchi and Gabriele Santin, University of Padova, Italy 10:25-10:45 Meshless Vector Field Approximation with Radial Basis Functions Edward Fuselier, High Point University, USA; Grady B. Wright, Boise State University, USA
Room:151 DE
Organizer: Donna Calhoun Boise State University, USA
Organizer: Carsten Burstedde Universität Bonn, Germany 9:10-9:30 Parallel, Adaptive, Multi Block Methods for Cartesian Grids using ForestClaw Donna Calhoun, Boise State University, USA; Carsten Burstedde, Universität Bonn, Germany 9:35-9:55 Progress in Parallel Adaptive Methods for Storm Surge Forecasting Kyle T. Mandli, Columbia University, USA 10:00-10:20 Using Explicit Filtering and Reconstruction to Improve LargeEddy Simulation of the Atmosphere on Adaptive Grids Lauren Goodfriend and Fotini Katopodes Chow, University of California, Berkeley, USA; Marcos Vanella and Elias Balaras, George Washington University, USA 10:25-10:45 Local Time Stepping for Parallel Adaptive Mesh Refinement Simulation Jeremy E. Kozdon and Lucas Wilcox, Naval Postgraduate School, USA
2015 SIAM Conference on Computational Science and Engineering
69
Sunday, March 15
Sunday, March 15
Sunday, March 15
MS103
IP4
PD2 Panel
High- and Low-order Finite Element Software for the Future - Part I of II 9:10 AM-10:50 AM
Extreme-scale Multigrid in Space and Time 11:20 AM-12:05 PM
Ask The Program Manager: Funding 12:15 PM-1:30 PM
Room:355
Room:355 D
Room:151 G
Chair: Irad Yavneh, Technion Israel Institute of Technology, Israel
Chair: Hank Childs, Lawrence Berkeley National Laboratory and University of Oregon, USA
For Part 2 see MS128 As scientific problems become larger, computational platforms become more diverse and the complexity of finite element software increases, it is essential to develop software in a sustainable way. The role of software design and the need to effectively manage the development process is therefore more important than ever before. This minisymposium will discuss experiences of how high- and low-order FEM software can be designed, developed and maintained to achieve maximum performance while remaining robust, rigorously tested and able to grow and adapt with the changing needs of the research environment, improvements in the methods and the evolving hardware landscape.
Organizer: David Moxey
Multigrid methods are important techniques for efficiently solving huge linear systems and they have already been shown to scale effectively on millions of cores. Future exascale architectures will require solvers to exhibit even higher levels of concurrency (1B cores), minimize data movement, exploit machine heterogeneity, and demonstrate resilience to faults. While considerable research and development remains to be done, multigrid approaches are ideal for addressing these challenges. In this talk, we will discuss efforts to develop extreme-scale multigrid, including a new parallel time integration approach that has the potential for significant speedups over standard time stepping.
Imperial College London, United Kingdom
Robert Falgout
Organizer: Chris Cantwell
Lawrence Livermore National Laboratory, USA
Imperial College London, United Kingdom
Organizer: Spencer Sherwin Imperial College London, United Kingdom 9:10-9:30 Architecting Spectral/HP Element Codes for Modern Hardware Chris Cantwell and David Moxey, Imperial College London, United Kingdom; Mike Kirby, University of Utah, USA; Spencer Sherwin, Imperial College London, United Kingdom 9:35-9:55 What Makes Computational Open Source Libraries Successful? Timo Heister, Clemson University, USA; Wolfgang Bangerth, Texas A&M University, USA 10:00-10:20 Applying Object-oriented Programming to PDE Solutions Hugh Blackburn, Monash University, Australia 10:25-10:45 Efficient Multithreading Algorithms for High-Order TensorProduct Finite Elements Timothy Warburton, David Medina, and Rajesh Gandham, Rice University, USA
Coffee Break 10:50 AM-11:20 AM Room:255
Lunch Break 12:05 PM-1:30 PM Attendees on their own
Chair: Vincent Heuveline, Heidelberg University, Germany Program managers from government agencies receive many requests for research funding. What are they really seeking? What makes a research proposal stand out? How can you build a research program that is attractive to these agencies? How can you conduct your research to make the biggest impact and increase your chances of future funding? What opportunities are presently available? We address all these questions and more as a part of this panel discussion. Panelists To Be Announced
70
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS104
MS105
Minitutorial: Python Visual Analytics for Big Data Part II of II 1:30 PM-3:10 PM
Featured Minisymposium: Distributed Methods for Optimization 1:30 PM-3:10 PM
Computational Engineering (BGCE) Student Paper PrizePart II of II 1:30 PM-3:10 PM
Room:355
Room:355 A
Room:250 A
For Part 1 see MT1 Chair: Jonathan Woodring, Los Alamos National Laboratory, USA
Recent advances in data collection, collaborative filtering, and wireless technology have lead to an unprecedented growth in demand for distributed optimization problems with huge data sets. The resulting optimization problems are characterized by distributed and uncertain information, necessitating computations to be done in a non-traditional environment, with imperfect information, over a communication network, and most importantly without a central entity that has an access to the whole information. This minisymposium focuses on most recent optimization techniques dealing with large data sets and distributed components over possibly uncertain networks.
For Part 1 see MS80 The 5th Bavarian Graduate School in Computational Engineering (BGCE) Student Paper Prize will be awarded at the 2015 SIAM CS&E Conference for outstanding student work in the field of Computational Science and Engineering. Eligible for the prize will be undergraduate and graduate students prior to receiving their PhD. Candidates are required to summarize their work in a short paper of at most 4 pages. The prize finalists will present their work in this minisymposium. The prize award announcement will be scheduled at one of the last days of the conference.
Sunday, March 15
MT2
Python is a powerful development, computational, and programming environment due to the wide variety of libraries developed for it, and importantly, the enthusiastic, active development and user community. One of the areas where Python excels is visualization and analysis of big data, due to several high-quality modules for both simple and advanced visual analytics. This tutorial will cover the following bigdata visualization capabilities in Python: interactive plotting with IPython, matplotlib, and databases, building web visualizations with Bokeh, and Python integration with VTK and ParaView. Additional information will be provided on mapreduce and NoSQL capabilities. This tutorial is intended for intermediate-level participants who have a basic understanding of the Python language and development environment (i.e., the student ought to have an understanding of native (and ideally numpy) data structures, file I/O, and is able to develop and run simple programs). Beginner participants are welcome, but Python fundamentals, such as language constructs, “hello world,” and program execution will not be covered in this tutorial. Topic C: Bokeh using Blaze Andy R. Terrel, Continuum Analytics, USA Topic D: VTK + ParaView using HDF5 & NetCDF Patrick O’Leary, Kitware, Inc., USA
Organizer: Wotao Yin University of California, Los Angeles, USA 1:30-1:50 Distributed Optimization in Directed Graphs: Push-Sum Based Algorithms Angelia Nedich and Alexander Olshevsky, University of Illinois at UrbanaChampaign, USA 1:55-2:15 Distributed Optimization in Undirected Graphs: Gradient and EXTRA Algorithms Wotao Yin, University of California, Los Angeles, USA; Qing Ling and Kun Yuan, University of Science and Technology of China, China 2:20-2:40 On the O(1/k) Convergence of Asynchronous Distributed Alternating Direction Method of Multipliers Ermin Wei and Asuman Ozdaglar, Massachusetts Institute of Technology, USA 2:45-3:05 Blessing of Scalability: A Tractable Dual Decomposition l-0 Approach for Large Graph Estimation Mengdi Wang, Princeton University, USA
Organizer: Tobias Neckel Technische Universität München, Germany
Organizer: Hans-Joachim Bungartz Technische Universität München, Germany
Organizer: Ulrich J. Ruede University of Erlangen-Nuremberg, Germany Speakers To Be Announced
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
MS106 High-dimensional Approximation and Integration: Analysis and Computation - Part II of V 1:30 PM-3:10 PM Room:250 B For Part 1 see MS81 For Part 3 see MS133 This minisymposium focuses on the fundamental problem of how to accurately approximate solutions of both forward and inverse high-dimensional functions. Predicting the behavior of complex phenomena relies on constructing solutions in terms of high dimensional spaces, particularly in the case when the input data (coefficients, forcing terms, initial and boundary conditions, geometry) are affected by large amounts of uncertainty. The resulting explosion in computational effort is a symptom of the curse of dimensionality and this symposium aims at exploring breakthroughs in sparse sampling and representations, nonlinear and greedy approximations, compressed sensing, multilevel methods, and ”best-N-term” approximations.
Organizer: Clayton G. Webster Oak Ridge National Laboratory, USA
Organizer: Guannan Zhang Oak Ridge National Laboratory, USA
Organizer: Max Gunzburger Florida State University, USA
Organizer: Albert Cohen Université Pierre et Marie Curie, France 1:30-1:50 Polynomial Approximation of Random PDEs by Discrete Least Squares Fabio Nobile, EPFL, Switzerland; Raul F. Tempone, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Giovanni Migliorati, EPFL, Switzerland 1:55-2:15 High Dimensional Interpolation for Non Intrusive Treatment of Parametric PDEs Abdellah Chkifa and Albert Cohen, Université Pierre et Marie Curie, France; Christoph Schwab, ETH Zürich, Switzerland continued in next column
2:20-2:40 Quasi-optimal Polynomial Approximation of PDEs with Linear and Nonlinear Stochastic Coefficients Clayton G. Webster, Hoang A. Tran, and Guannan Zhang, Oak Ridge National Laboratory, USA; Ronald DeVore, Texas A&M University, USA 2:45-3:05 Combining Sparsity and Smoothness for Function Interpolation Rachel Ward, University of Texas at Austin, USA; Holger Rauhut, RWTH Aachen University, Germany
71
Sunday, March 15
MS107 Resilience in Numerical Simulations and Algorithms at Extreme Scale: Part II of IV - Runtime 1:30 PM-3:10 PM Room:250 C For Part 1 see MS82 For Part 3 see MS134 The advent of extreme scale machines will require the use of parallel resources at an unprecedented scale, probably leading to a high rate of hardware faults. Handling fully these faults at the computer system level may have a prohibitive cost. High performance computing applications that aim at exploiting all these resources will thus need to be resilient, in this minisymposium different complementary approaches and methods will be presented to possibly address this key aspect of extreme scale computing.
Organizer: Keita Teranishi Sandia National Laboratories, USA
Organizer: Luc Giraud INRIA, France
Organizer: Michael Heroux Sandia National Laboratories, USA
Organizer: Emmanuel Agullo INRIA, France 1:30-1:50 Runtime Systems for Fault Tolerant Computing Marc Casas, Barcelona Supercomputing Center, Spain 1:55-2:15 Fenix: A Framework for Online Failure Recovery for Scientific Simulations Towards Exascale Marc Gamell, Rutgers University, USA; Daniel Katz, Argonne National Laboratory, USA; Keita Teranishi, Hemanth Kolla, and Jacqueline Chen, Sandia National Laboratories, USA; Scott Klasky, Oak Ridge National Laboratory, USA; Manish Parashar, Rutgers University, USA 2:20-2:40 Understanding the Impact of Transient Faults at the Application Level in HPC Ignacio Laguna, Lawrence Livermore National Laboratory, USA 2:45-3:05 DHARMA: Distributed asyncHronous Adaptive Resilient Management of Applications Hemanth Kolla, Janine C. Bennett, Jeremiah Wilke, Nicole Slattengren, Keita Teranishi, and John Floren, Sandia National Laboratories, USA
72
Sunday, March 15
MS108 Preconditioning and Iterative Methods for Linear Systems - Part II of II 1:30 PM-3:10 PM Room:250 D For Part 1 see MS83 The solution of large, sparse linear systems remains the limiting component of many problems. Iterative solvers are often the methods of choice for such systems, and effective preconditioning is usually necessary to achieve an acceptable speed of convergence. In this minisymposium we focus on recent developments in preconditioning, the convergence of solvers and the application of iterative methods.
Organizer: Jennifer Pestana University of Manchester, United Kingdom
Organizer: Andy Wathen University of Oxford, United Kingdom 1:30-1:50 Exploiting Tropical Algebra in the Construction of Preconditioners James Hook, Jennifer Pestana, and Francoise Tisseur, University of Manchester, United Kingdom 1:55-2:15 The Solution of Lyapunov Equations with Nonnormal Coefficients Mark Embree, Virginia Tech, USA; Jonathan Baker, Rice University, USA; John Sabino, The Boeing Company, USA 2:20-2:40 The Surprising Robustness of the Tracemin Eigensolver to Incorrect Linear Solves Alicia Klinvex, Purdue University, USA; Michael Heroux, Sandia National Laboratories, USA; Ahmed Sameh, Purdue University, USA; Mark Hoemmen, Sandia National Laboratories, USA 2:45-3:05 Numerical Solution of PDEs Posed on Graphs Michele Benzi, Emory University, USA
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
MS109 Error Analysis and Scalability of UQ Methodologies for Inverse Problems - Part II of II 1:30 PM-3:10 PM
Sunday, March 15
MS110 Advanced Finite Element Methods for Nonlinear Materials and Fluids Part II of III 1:30 PM-3:10 PM
Room:250 E
Room:250 F
For Part 1 see MS84 The application of novel uncertainty quantification (UQ) methodologies to solve stochastic inverse problems for a physical system has improved the predictive capabilities of many computationally complex models. However, a full error analysis and study of scalability for the combined UQ method and computational model is often lacking. This session addresses the issues of error analysis and scalability of several UQ methods applied to stochastic inverse problems associated with specific complex physical systems. Both a priori and a posteriori error analyses of deterministic and stochastic sources of error are considered along with computational challenges in scaling such analyses.
For Part 1 see MS85 For Part 3 see MS137 Numerical modeling of materials and fluids with nonlinear behavior plays an increasing role in computational science and engineering, particularly in the high-fidelity simulation of elastic and plastic deformations, magnetosensitive materials, liquid crystals, biological tissues and blood flow, among others. Despite a long history of efforts, there is still a need for the development of advanced finite-element methods with specific properties such as accurate representation of stresses, forces, and other quantities depending on the application. In many of these applications, mesh-adaptive implementations and efficient solvers are an important issue due to the size and complexity of the problems.
Organizer: Troy Butler
Organizer: James H. Adler
University of Colorado, Denver, USA
Tufts University, USA
Organizer: Steven Mattis
Organizer: Gerhard Starke
University of Texas at Austin, USA
University of Duisburg-Essen, Germany
1:30-1:50 A Scalable MeasureTheoretic Approach to the Stochastic Inverse Problem for Groundwater Contamination Steven Mattis and Clint Dawson, University of Texas at Austin, USA; Troy Butler, University of Colorado, Denver, USA
Organizer: Thomas Manteuffel
1:55-2:15 A Scalable Computational Framework for Estimating Model Discrepancy Nishant Panda, Colorado State University, USA 2:20-2:40 Quantifying Error in An Inadequate Model for Flow in a Porous Media Teresa Portone, Damon Mcdougall, Todd Oliver, and Robert D. Moser, University of Texas at Austin, USA 2:45-3:05 Error Decomposition and Adaptivity for Response Surface Approximations with Application to Bayesian Inference Corey M. Bryant, Serge Prudhomme, and Todd Oliver, University of Texas at Austin, USA; Tim Wildey, Sandia National Laboratories, USA
University of Colorado Boulder, USA 1:30-1:50 Hybrid FOSLS/ll* for Nonlinear Systems of PDEs Thomas Manteuffel, Kuo Liu, Lei Tang, and John Ruge, University of Colorado Boulder, USA; Chad Westphal, Wabash College, USA 1:55-2:15 Nested Iteration and Adaptive Finite Elements for Ice Sheet Models Jeffery M. Allen and Thomas Manteuffel, University of Colorado Boulder, USA; Harihar Rajaram, University of Colorado, USA 2:20-2:40 Parametric Mixed Finite Elements for Two-Phase Flow Interface Problems Fleurianne Bertrand, Gerhard Starke, and Steffen Münzenmaier, University of Duisburg-Essen, Germany 2:45-3:05 A Least Squares Finite Element Method for Coupled Surface/ Subsurface Flows Steffen Münzenmaier, University of DuisburgEssen, Germany
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
MS111 Adjoint Methods for Unsteady and Chaotic PDEs and Large-scale Optimization and Control Part II of II 1:30 PM-3:10 PM Room:251 A For Part 1 see MS86 We expose a growing intersection of optimization and control using largescale, unsteady PDE-based simulations. This intersection has traditionally been small because most practitioners perform optimization and control using inexpensive, reduced order models; however, this is changing. Current-day simulations solving unsteady and chaotic PDEs, such as the Navier-Stokes equations, are leveraging advance in high performance computing. Petascale, and soon exascale, supercomputers are increasingly available to researchers, enabling them to attempt optimization and control directly using large-scale simulations. The resulting area of large-scale optimization and control, particularly with the aid of the unsteady adjoint method, is the theme of this proposed minisymposium.
Organizer: Qiqi Wang Massachusetts Institute of Technology, USA
Organizer: Nicolas R. Gauger Technische Universität Kaiserslautern, Germany
Organizer: Daniel J. Bodony University of Illinois at Urbana-Champaign, USA 1:30-1:50 Actuator and Sensor Placement for Controlling HighSpeed Jet Noise Mahesh Natarajan and Daniel J. Bodony, University of Illinois at UrbanaChampaign, USA 1:55-2:15 Using Imperfect Outputs and Derivatives in Large-scale Optimization Jason E. Hicken, Rensselaer Polytechnic Institute, USA
continued in next column
2:20-2:40 Least Squares Shadowing for Adjoint Calculation of Chaotic and Turbulent Pdes Patrick Blonigan and Qiqi Wang, Massachusetts Institute of Technology, USA 2:45-3:05 Parallel Bayesian Optimization of Massively Parallel Turbulent Flow Simulations Chaitanya Talnikar, Massachusetts Institute of Technology, USA
73
Sunday, March 15
MS112 Scientific Software Productivity at Extreme Scale 1:30 PM-3:10 PM Room:251 B There has been a recent renewed focus on software productivity for scientific applications, especially those that require performance at extreme scales (10k+ and GPU-accelerated cores). The goal of this minisymposium is to engage the broader computational science and engineering software community, to discuss productivity issues that arise when developing highperformance applications. The session will include an overview presentation of forums and efforts focusing on productivity (including a recently funded initiative within DOE); two science application presentations, oriented towards software productivity considerations; and an open facilitated session to gather SIAM CSE community concerns and priorities.
Organizer: Hans Johansen Lawrence Berkeley National Laboratory, USA
Organizer: Lois Curfman McInnes Argonne National Laboratory, USA 1:30-1:50 Overview: Software Productivity Challenges for Extreme Scale Science Lois Curfman McInnes, Argonne National Laboratory, USA; Hans Johansen, Lawrence Berkeley National Laboratory, USA; Michael Heroux, Sandia National Laboratories, USA 1:55-2:15 Software Productivity Application: Integrated Modeling for Fusion Energy David E. Bernholdt, Wael R. Elwasif, and Donald B. Batchelor, Oak Ridge National Laboratory, USA 2:20-2:40 Software Productivity Challenges in Environmental Applications David Moulton and Ethan T. Coon, Los Alamos National Laboratory, USA; Carl Steefel, Lawrence Berkeley National Laboratory, USA 2:45-3:05 Software Productivity Community Input: Concerns and Priorities Jeffrey C. Carver, University of Alabama, USA; Michael Heroux, Sandia National Laboratories, USA
74
Sunday, March 15
MS113 Fluid Transport Dynamics in Biology and Medicine 1:30 PM-3:10 PM Room:251 C Part of the SIAM Workshop Celebrating Diversity Fluid dynamics, especially blood flow, govern many physiological processes. The function of biological entities and the success of clinical treatments can be affected by the fluid environment. Understanding the transport mechanisms from vessels into the surrounding medium is also critical in the design of drug delivery vehicles and treatment protocols. This session will highlight mathematical and computational methods for simulating important flow-dependent processes, including biological and therapeutic applications.
Organizer: Erica J. Graham North Carolina State University, USA
Organizer: Ami Radunskaya Pomona College, USA
Organizer: Raegan Higgins Texas Tech University, USA
Organizer: Susan E. Minkoff University of Texas at Dallas, USA
Organizer: Stephen Wirkus Arizona State University, USA 1:30-1:50 Modeling the Effects of Flow on Anticoagulant Therapy Erica J. Graham, North Carolina State University, USA; Lisette dePillis, Harvey Mudd College, USA; Kaitlyn Hood, University of California, Los Angeles, USA; Yanping Ma, Loyola Marymount University, USA; Julie Simons, Tulane University, USA; Ami Radunskaya, Pomona College, USA 1:55-2:15 Navier Slip Condition for Viscous Fluids on a Rough Boundary Silvia Jimenez Bolanos, Colgate University, USA; Bogdan M. Vernescu, Worcester Polytechnic Institute, USA 2:20-2:40 Cooperative Swimming in Viscous Environments Julie Simons, Lisa J. Fauci, and Ricardo Cortez, Tulane University, USA 2:45-3:05 Taming Targeted Drug Delivery: a Mathematical Model of Triggered Drug Delivery Across the Blood-Brain Barrier Ami Radunskaya, Pomona College, USA
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
MS114 AWM Meeting - Workshop: Research Talks by Recent Ph.D.s: Mathematical Modeling and Highperformance Computing for Multi-physics and Multiscale Problems. Part II of II 1:30 PM-3:10 PM Room:251 D For Part 1 see MS89 Sponsored by Association for Women in Mathematics (AWM) This minisymposium highlights selected research talks by female recent Ph.D.s in developing multiscale multiphysics models, computational methodologies, and highperformance computing. Advances in science and engineering—in disciplines ranging from health to energy and the environment to defense—rely on predictive measurement and analysis of multiscale multiphysics systems. Such advances are essential for scientific discovery, engineering design and policy making. The participants of the minisymposium will discuss advanced computational algorithms and simulation strategies for solving modern and nextgeneration systems that require analyzing complex physical phenomena such as fluidsolid interactions, mechanics of biological materials, moleculer motions, charge transport, ocean and cloud dynamics, and oil recovery process. The topics include physical modeling, mathematical analysis, numerical analysis, algorithms, implementation, performance, and scalability. This minisymposium is part of the Association for Women in Mathematics(AWM) workshop that will promote cross-fertilization of ideas among women scientists in multiple disciplines.
Organizer: MiSun Min Argonne National Laboratory, USA
Organizer: Fengyan Li Rensselaer Polytechnic Institute, USA
Organizer: Jingmei Qiu University of Houston, USA 1:30-1:50 A Study of the Entanglement in Polymer Melts Eleni Panagiotou, University of California, Santa Barbara, USA continued in next column
1:55-2:15 A Fast Explicit Operator Splitting Method for a Multi-scale Underground Oil Recovery Model Ying Wang, University of Oklahoma, USA 2:20-2:40 Computational Study of Dynamics and Transport in VortexDipole Flows Ling Xu, Georgia State University, USA 2:45-3:05 A Stabilized Explicit Scheme for Coupling Fluid-structure Interactions Yue Yu, Lehigh University, USA
2015 SIAM Conference on Computational Science and Engineering
75
Sunday, March 15
Sunday, March 15
Sunday, March 15
MS115
MS116
MS117
Software and Methods for Spatial Modeling and Simulation in Systems Biology - Part II of II 1:30 PM-3:10 PM
Parametric Model Reduction and Inverse Problems - Part II of IV 1:30 PM-2:45 PM
Applied Mathematics and Computation in the Pharmaceutical Industry Part II of II 1:30 PM-2:45 PM
Room:251 E
For Part 1 see MS91 For Part 3 see MS143 In recent years model reduction techniques have been used to efficiently solve forward problems given, for instance, as PDEs, dynamical systems or complex networks. In this minisymposium we want explore applications of model reduction in the context of inverse problems.
For Part 1 see MS90 Simulation of spatially inhomogeneous biological systems is a challenging problem that has motivated the development of various software packages. Simulation challenges include the multiphysics/ multiscale nature of the systems and the need to capture stochastic effects. Modeling challenges include specification of the reaction systems, geometry and mesh generation for complex and dynamic shapes. Software issues include testing and verification, user community issues, and dissemination and promotion. Also, there are the issues of model sharing, validation and reproducibility of results. This minisymposium brings together representatives of different software packages from the computational systems biology community to discuss these challenges.
Organizer: Andreas Hellander Uppsala University, Sweden 1:30-1:50 Interactive, Distributed Spatial Stochastic Simulation with PyURDME and MOLNs Brian Drawert, University of California, Santa Barbara, USA 1:55-2:15 Gepetto/OpenWorm Stephen Larson, OpenWorm.org, USA 2:20-2:40 Reaction-diffusion and Electrical Signaling in Neurons (Rdesigneur): a System for Multiscale Modeling in MOOSE Upinder Bahlla, National Center for Biological Sciences, India 2:45-3:05 Stochastic Simulation at Your Service Linda R. Petzold, University of California, Santa Barbara, USA
Room:251 F
Room:254 A
Organizer: Christopher A. Beattie
For Part 1 see MS92 The pharmaceutical industry continues to rapidly develop analytical capabilities tackling scientific, engineering, manufacturing and market challenges. We highlight current mathematical and computational frontiers and areas of interest to the industry in this minisymposium. Specific areas include modeling and simulation, optimization, image processing and data mining/modeling. Modeling and simulation is used in both scientific and business applications within the industry. We highlight disease and treatment models as well as models underlying objective functions in optimization. Data mining/ modeling is now driven by large amounts of biomarker data varying from next generation sequencing to ultra-high resolution imaging.
Virginia Tech, USA
Organizer: Jeffrey Saltzman
Organizer: Serkan Gugercin
AstraZeneca, USA
Virginia Tech, USA
1:30-1:50 Imaging Biomarkers in Biopharmaceutical Industry Belma Dogdas, Merck & Co., Inc., USA
Organizer: Eldad Haber University of British Columbia, Canada
Organizer: Lars Ruthotto Emory University, USA
Organizer: Eric De Sturler Virginia Tech, USA
Organizer: Misha E. Kilmer Tufts University, USA
1:30-1:50 From Data to Prediction Via Reduced Parameter-to-Observable Maps: Applications to Antarctic Ice Sheet Flow Omar Ghattas and Toby Isaac, University of Texas at Austin, USA; Noemi Petra, University of California, Merced, USA; Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA 1:55-2:15 Combined State and Parameter Reduction for the Inversion of Functional Neuroimaging Data Christian Himpe, University of Münster, Germany; Mario Ohlberger, University of Muenster, Germany 2:20-2:40 Convergence of Inverse Problems using Reduced Order Models Eric De Sturler, Virginia Tech, USA
1:55-2:15 Imaging Genomics for Pharmaceutical Applications Sangeetha Somayajula and Chandni Valiathan, Merck & Co., Inc., USA 2:20-2:40 Simulation-Based Analysis of Complex Decision Options in Pharmaceutical Research and Development Otto Ritter, Independent Consultant, USA
76
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS118
MS119
MS120
Computational Applications Performing at Petascale Level and Beyond Part II of II 1:30 PM-3:10 PM
Student Days: SIAM Student Chapter Presentations Part II of II 1:30 PM-3:10 PM
Computational Methods for Kinetic Equations and Related Models - Part V of V 1:30 PM-3:10 PM
Room:254 C
Room:150 AB
Room:254 B
For Part 1 see MS94 SIAM Student Chapter presentations.
For Part 4 see MS95 Kinetic descriptions play an important role in a variety of physical, biological, and even social applications. Unfortunately, the large phase space associated with the kinetic description has in the past made simulations impractical in most settings. However, recent advances in computer resources and numerical algorithms are making kinetic models more tractable, and this trend is expected to continue in the future. The purpose of this minisymposium is to report on the continuing progress on numerical analysis and computational science for kinetic equations. It brings together researchers from applied mathematics, computational science, physics, and engineering communities to discuss their work and exchange ideas.
Sunday, March 15
For Part 1 see MS93 Large-scale simulations are required to validate and then exercise complex multiscale models such as weather forecasting, computational astronomy, or industrial burners. This minisymposium highlights the impact of petascale computing on scientific discoveries by presenting the recent work of several research teams that develop and employ applications to tackle such challenging problems using full capabilities of modern leadership-class supercomputers. Discussions emphasize both the scientific impact of the related work and the challenges associated with computing at such scale.
Organizer: Bilel Hadri King Abdullah University of Science & Technology (KAUST), Saudi Arabia
Organizer: Hatem Ltaief King Abdullah University of Science & Technology (KAUST), Saudi Arabia 1:30-1:50 What Are the Priorities Beyond Petascale Computing? Thomas C. Schulthess, ETH Zürich, Switzerland 1:55-2:15 Petascale Medical Simulations Michael Resch, University of Stuttgart, Germany; Ralf Schneider, High Performance Computing Center Stuttgart, Germany 2:20-2:40 Petascale Simulation of Hurricane Sandy Using WRF Weather Model on Cray XE6 Blue Waters Pete Johnsen, Cray, Inc., USA; Mark Starka, National Center for Supercomputing Applications (NCSA), USA; Melvyn Shapiro, Alan Norton, and Thomas Galarneau, National Center for Atmospheric Research, USA
Organizer: Rachel Levy Harvey Mudd College, USA 1:30-1:41 Sublinear Preconditioners for the 2D Helmholtz Equation Leonardo Zepeda-Núñez and Laurent Demanet, Massachusetts Institute of Technology, USA 1:46-1:57 Solving the Heat Equation with Wavelets Anne Reinarz and Alexey Chernov, University of Reading, United Kingdom 2:02-2:13 Analysis of a Heterogeneous Multiscale Method for Poroelasticity Paul M. Delgado, Vinod Kumar, and Son Young Yi, University of Texas at El Paso, USA 2:18-2:29 Experimental Analysis of the Performance of GeoClaw, AnuGA and SurfWB-UC Numerical Models for the Simulation of Tsunami Inundation Phenomena José Galaz and Rodrigo Cienfuegos, Pontificia Universidad Católica de Chile, Chile
Organizer: Martin Frank RWTH - Aachen University of Technology, Germany
Organizer: Cory Hauck Oak Ridge National Laboratory, USA
Organizer: Ryan G. McClarren Texas A&M University, USA
Organizer: Jingmei Qiu
2:34-2:45 Stabilization in Relation to Wavenumber in HDG Methods Nicole Olivares and Jay Gopalakrishnan, Portland State University, USA; Liang Li, University of Electronic Science and Technology of China, China; Ronan Perrussel, Universite de Toulouse, France
University of Houston, USA
2:50-3:01 Incompressible Flow and (Stabilised) Mixed Finite Element Methods on Highly Stretched Meshes Andreas Wachtel, University of Strathclyde, United Kingdom; Mark Ainsworth, Brown University, USA; Gabriel R. Barrenechea, University of Strathclyde, United Kingdom
1:55-2:15 A Fast Conservative Spectral Solver for the Nonlinear Boltzmann Collision Operator Jeffrey Haack, University of Texas at Austin, USA; Jingwei Hu, Purdue University, USA; Irene M. Gamba, University of Texas at Austin, USA
Organizer: Jeffrey A. Willert Los Alamos National Laboratory, USA 1:30-1:50 Kinetic Equation in a Bounded Domain Qin Li, California Institute of Technology, USA
2:45-3:05 Petascale Simulations of Cloud Cavitation Collapse Costas Bekas, IBM Research, USA continued on next page
2015 SIAM Conference on Computational Science and Engineering
2:20-2:40 High-Order SemiLagrangian Discontinuous Galerkin Methods for Kinetic Plasma Models David C. Seal, Michigan State University, USA; James A. Rossmanith, Iowa State University, USA; Andrew J. Christlieb, Michigan State University, USA 2:45-3:05 A Monte Carlo Method with Negative Particles for General Binary Collisions and Application to Coulomb Collisions Bokai Yan and Russel Caflisch, University of California, Los Angeles, USA
Sunday, March 15
MS121 Frameworks, Algorithms, and Scalable Technologies for Parallel Unstructured Mesh Workflows - Part II of II 1:30 PM-3:10 PM Room:258 For Part 1 see MS97 This minisymposium presents recent advances in the development of parallel unstructured mesh methods and highlights their use in applications. These technologies are designed and implemented to support reliable simulation technologies that run on today’s leadership-class parallel systems with specific attention to next generation and emerging architectures that will bridge the peta-to-exascale gap. Unstructured meshing technologies presented in these sessions will emphasize memory-efficiency through array based data structures and adaptivity, increased fidelity through high-order methods, and parallelism in mesh generation, adaptation, and in-memory workflows. Many of the minisymposium presentations are associated with the DOE SciDAC FASTMath institute (http://www.fastmathscidac.org/).
Organizer: Cameron Smith Rensselaer Polytechnic Institute, USA
Organizer: Mark S. Shephard Rensselaer Polytechnic Institute, USA 1:30-1:50 Parallel Meshing Technologies for Large Scale Adaptive Simulations Saurabh Tendulkar, Mark Beall, and Rocco Nastasia, Simmetrix, Inc., USA; Mark S. Shephard, Rensselaer Polytechnic Institute, USA 1:55-2:15 OCC-Based Meshing for RGG Applications Using MeshKit Rajeev Jain, Argonne National Laboratory, USA; Jacob Becker, Kitware, Inc., USA; Vijay Mahadevan, Argonne National Laboratory, USA; Patrick Shriwise, University of Wisconsin, Madison, USA; Robert O’Bara, Kitware, Inc., USA; Iulian Grindeanu, Argonne National Laboratory, USA
continued in next column
77
2:20-2:40 Parallel Mesh Curving and Adaptation with High-Order Surface Continuity for High-Order Finite Element Simulations Qiukai Lu, Dan A. Ibanez, and Mark S. Shephard, Rensselaer Polytechnic Institute, USA 2:45-3:05 High-Order Surface Reconstruction with Applications in Parallel Meshing and Finite Element Solvers Xiangmin Jiao, State University of New York, Stony Brook, USA; Navamita Ray, Argonne National Laboratory, USA; Cao Lu and Xinglin Zhao, Stony Brook University, USA
78
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS123
MS124
Computational Approaches and Multi-scale Modeling of Complex Fluids - Part II of II 1:30 PM-3:10 PM
High-Order Methods for Computational Fluid Dynamics - Part II of IV 1:30 PM-3:10 PM
Large Scale Phasefield Simulations of Solidification Processes 1:30 PM-3:10 PM
Room:150 DE
Room:259
Room:260 A
For Part 1 see MS96 Complex fluids, which are suspensions of particles in a viscous liquid, can exhibit both viscous and elastic response to external forcing. Capturing the exact interplays between the particles and the surrounding fluid in a continuum modeling presents significant challenges as hydrodynamics interactions happen on short and long scales. Added difficulties arise when such suspensions interact with boundaries, where peculiar behavior has been experimentally observed, or when the surrounding fluid itself is viscoelastic, as it is the case of most biological liquid. This minisymposium’s goal is to explore recent advances, especially on the computational front, on both topics.
For Part 1 see MS98 For Part 3 see MS150 Novel contributions in the field of highorder numerical methods for computational fluid dynamics. Presentations will explore new high-order methods, benchmarking of existing schemes, and applications to turbulent flow problems.
In this minisymposium we focus on the formation of patterns during the solidification process of binary and ternary eutectic systems using phasefield simulations. Depending on several different material and process parameters, the occurring patterns are changing. These patterns have a significant influence on the mechanical properties of the component. For the prediction of realistic and nonideal systems and to minimize the boundary effects, large simulation domains are required. To solve the phase-field evolution equations coupled with the evolution of concentration and temperature, in large domains, highly parallel and optimized algorithms are needed. For this some of the largest, currently available, computers are used.
Sunday, March 15
MS122
Organizer: Christel Hohenegger University of Utah, USA
Organizer: Enkeleida Lushi Brown University, USA 1:30-1:50 Computational Models of Cilia and Flagella in a Brinkman Fluid Karin Leiderman, University of California, Merced, USA; Sarah D. Olson, Worcester Polytechnic Institute, USA 1:55-2:15 Accurate Simulations of Complex Fluid Flow in Domains with Smooth Boundaries Using Fft-Based Spectral Methods David Stein and Becca Thomases, University of California, Davis, USA 2:20-2:40 Fluid Coupling in Continuum Modeling of Microtubule Gliding Assays Tamar Shinar and Steven Cook, University of California, Riverside, USA; Christel Hohenegger, University of Utah, USA 2:45-3:05 Modeling Active Flows and Stress Generation in MicrotubuleMotor Networks Robert Blackwell, Meredith Betterton, and Matthew Glaser, University of Colorado, USA; Michael J. Shelley and Tony Gao, Courant Institute of Mathematical Sciences, New York University, USA
Organizer: Brian C. Vermeire Imperial College London, United Kingdom
Organizer: Antony Jameson Stanford University, USA
Organizer: Peter E. Vincent Imperial College London, United Kingdom 1:30-1:50 Further Developments in the Flux Reconstruction Method Antony Jameson, Stanford University, USA 1:55-2:15 Spectral Difference Method for Large Eddy Simulation Using NonConforming and Sliding Meshes Bin Zhang and Chunlei Liang, George Washington University, USA 2:20-2:40 Turbulent Wall Modelling for High-Order Unstructured Methods on GPUs Brian C. Vermeire and Peter E. Vincent, Imperial College London, United Kingdom 2:45-3:05 High-Order Methods for Turbulent Flow Simulations on Deforming Domains Per-Olof Persson, University of California, Berkeley, USA
Organizer: Harald Koestler University of Erlangen-Nuremberg, Germany
Organizer: Martin Bauer University of Erlangen-Nuremberg, Germany 1:30-1:50 Massively Parallel Phase-Field Simulations using HPC Framework waLBerla Martin Bauer, Harald Koestler, and Ulrich J. Ruede, University of ErlangenNuremberg, Germany 1:55-2:15 Large Scale and Massive Parallel Phase-field Simulations of Pattern Formations in Ternary Eutectic Systems Johannes Hötzer, Hochschule Karlsruhe Technik und Wirtschaft, Germany; Marcus Jainta, Philipp Steinmetz, and Britta Nestler, Karlsruhe Institute of Technology, Germany
continued on next page
2015 SIAM Conference on Computational Science and Engineering
79
2:20-2:40 Large-scale Multi-PhaseField Simulation of Abnormal Polycrystalline Grain Growth using TSUBAME2.5 GPU-Supercomputer Akinori Yamanaka and Masashi Okamoto, Tokyo University of Agriculture and Technology, Japan; Takashi Shimokawabe and Takayuki Aoki, Tokyo Institute of Technology, Japan
Sunday, March 15
Sunday, March 15
MS125
MS126
Analysis of Noisy Networks in Theory and Practice Part II of II 1:30 PM-3:10 PM
2:45-3:05 Multi-Gpu Phase-Field Lattice Boltzmann Simulations for Growth and Moving of Binary Alloy Dendrite Tomohiro Takaki and Roberto Rojas, Kyoto Institute of Technology, Japan; Takashi Shimokawabe and Takayuki Aoki, Tokyo Institute of Technology, Japan
Room:260 B
Advances in Radial Basis Function and Other Meshfree Methods Part V of V 1:30 PM-3:10 PM
For Part 1 see MS100 Network analysis is an elegant, graph-based method for studying complex systems of interacting entities. However, real-world observed systems inherently include some discrepancy from reality. For example, measurements may not be exact, as not all relations are observed. Since traditional graph theory deals with exact mathematics, one of the most challenging aspects of network analysis is accounting for this noise. This minisymposium will include talks from application experts on how they handle noise, and presentations about the effect of noise in networks generally. We conclude with a discussion on how the general theory and application-specific methods can benefit from each other.
Room:151 AB
Organizer: Sanjukta Bhowmick
Organizer: Varun Shankar
University of Nebraska, Omaha, USA
Organizer: Benjamin A. Miller Massachusetts Institute of Technology, USA 1:30-1:50 Pockets of Instability in Network Centrality Metrics Vladimir Ufimtsev and Sanjukta Bhowmick, University of Nebraska, Omaha, USA
For Part 4 see MS101 Meshfree methods have many inherent properties that make them useful for a variety of problems in science and engineering, from fitting data to numerically solving differential equations. Such methods offer geometric flexibility, non-uniform resolution, and advantageous trade-offs between accuracy and computational costs. This minisymposium focuses primarily on methods based on radial basis functions and other more general kernels. The talks will address recent advances in the application of these methods to large-scale problems in biology, geophysics, image processing, and finance, as well as theoretical advances in the methods themselves. University of Utah, USA
Organizer: Grady B. Wright Boise State University, USA 1:30-1:50 Kernel-Based Image Reconstruction Frank Filbir, Technische Universität München, Germany
1:55-2:15 Using Consensus to Inform Stochastic Graph Aggregation Layla Oesper, Brown University, USA; Rajmonda Caceres, Massachusetts Institute of Technology, USA
1:55-2:15 Reproducing Kernels in Parametric Partial Differential Equations Christian Rieger, Universität Bonn, Germany
2:20-2:40 Statistical Inference on Errorfully Observed Graphs Carey Priebe, Johns Hopkins University, USA; Daniel L. Sussman, Harvard University, USA; Minh Tang, Johns Hopkins University, USA; Joshua Vogelstein, Duke University, USA; Vince Lyzinski, Donniell Fishkind, Nam Lee, Youngser Park, and Avanti Athreya, Johns Hopkins University, USA
2:20-2:40 Meshfree Computations with SPH and Vortex Methods Prabhu Ramachandran, Indian Institute of Technology-Bombay, India
2:45-3:05 Discussion on Future Directions of Noisy Networks Sanjukta Bhowmick, University of Nebraska, Omaha, USA; Benjamin A. Miller, Massachusetts Institute of Technology, USA
2:45-3:05 A Numerical Study of the Accuracy of Divergence-Free Kernel Approximations Arthur Mitrano, Arizona State University, USA
80
Sunday, March 15
MS127
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
MS128
Poster Blitz
Parallel, Adaptive Methods for Logically Cartesian Meshes - Part II of II 1:30 PM-3:10 PM
High- and Low-order Finite Element Software for the Future - Part II of II 1:30 PM-3:10 PM
Room:151 DE
Room:151 G
For Part 1 see MS102 Adaptive refined Cartesian mesh methods, which started with the seminal paper by Berger-Oliger in 1984 have now been in development for just over 30 years. However, there are still many questions about the scalability of these methods. The meta-data needed to store sub-grid locations can become prohibitive, and dynamic refinement an present a challenging parallel problem. However recent developments in quad tree adaptation, as well as other approaches taken for Berger Oliger type refinement may overcome some of these problems. In this minisymposium, speakers have been asked to address the scalability of their adaptive frameworks, and more generally, share experiences using available packages, including Chombo, Paramesh, p4est, Overture, AMRClaw, GeoClaw and ForestClaw.
For Part 1 see MS103 As scientific problems become larger, computational platforms become more diverse and the complexity of finite element software increases, it is essential to develop software in a sustainable way. The role of software design and the need to effectively manage the development process is therefore more important than ever before. This minisymposium will discuss experiences of how high- and low-order FEM software can be designed, developed and maintained to achieve maximum performance while remaining robust, rigorously tested and able to grow and adapt with the changing needs of the research environment, improvements in the methods and the evolving hardware landscape.
Organizer: Donna Calhoun Boise State University, USA
Organizer: Carsten Burstedde Universität Bonn, Germany 1:30-1:50 Recent Developments in Forest-of-octrees AMR Carsten Burstedde, Universität Bonn, Germany; Toby Isaac and Omar Ghattas, University of Texas at Austin, USA; Donna Calhoun, Boise State University, USA 1:55-2:15 Runtimes and Autotuning and Hybrid, Oh My! Chombo Navigates the Waters of Exascale Brian Van Straalen, Lawrence Berkeley National Laboratory, USA 2:20-2:40 Exploring Astrophysical Flows with High-fidelity Large-scale Simulations Min Long, Boise State University, USA 2:45-3:05 An Implicit, High-Order Accurate, Incompressible NavierStokes Solver on Overlapping Grids Kyle Chand, Lawrence Livermore National Laboratory, USA
Organizer: David Moxey Imperial College London, United Kingdom
Organizer: Chris Cantwell Imperial College London, United Kingdom
Organizer: Spencer Sherwin Imperial College London, United Kingdom 1:30-1:50 Nek5000: An Environment for Scalable Algorithm Development and Production Simulations Paul F. Fischer, Argonne National Laboratory, USA 1:55-2:15 Anisotropic Mesh Adaptation for the Many-core Era Gerard J Gorman, Imperial College London, United Kingdom 2:20-2:40 H-to-P Efficiently: a Nektar++ Update on Comparisons of Cg and Hdg Mike Kirby, University of Utah, USA 2:45-3:05 Heterogeneous Computing with a Homogeneous Codebase Freddie Witherden and Peter E. Vincent, Imperial College London, United Kingdom
3:10 PM-4:30 PM Room:355
Sunday Poster Session & Reception PP1 Computational Fluid Dynamics 4:30 PM-6:30 PM Room:255 Lid Driven Cavity Simulations in 2d and 3d Using High Accurate Methods Badr Alkahtani, King Saud University, Saudia Arabia Numerical Study of Thin Viscoelastic Films on Substrates Valeria Barra and Shahriar Afkhami, New Jersey Institute of Technology, USA Flusepa - a Navier-Stokes Solver for Unsteady Problems with Bodies in Relative Motion : Toward a Task-Based Parallel Version over a Runtime System for Large Simulations Jean Marie Couteyen Carpaye and Jean Roman, INRIA, France; Pierre Brenner, Airbus Defence and Space, Germany A Novel Modeling Approach for Multiscale, Multiphysics Flow Derek A. Cline, University of Utah, USA Efficiency of an Adjoint Industrial CFD Code Zahrasadat Dastouri, Johannes Lotz, and Uwe Naumann, RWTH Aachen University, Germany Production of Dissipative Vortices by Solid Bodies in the Inviscid Limit of Incompressible Fluid Flows: Comparison Between Prandtl, NavierStokes and Euler Solutions Marie Farge, Ecole Normale Superieure, France; Romain Nguyen van yen and Matthias Waidmann, Freie Universitaet Berlin, Germany; Kai Schneider, AixMarseille Université, France; Rupert Klein, Freie Universitaet Berlin, Germany
continued on next page
2015 SIAM Conference on Computational Science and Engineering
Physics-Compatible Lagrangian Space and Time-Staggered Hydrodynamics Schemes Christophe Fochesato, Alexandra Claisse, and Antoine Llor, CEA, France A Numerical Study of Shock-Induced Cavity Collapse in a Solid Explosive James R. Gambino, Ashwani K. Kapila, Donald W. Schwendeman, and William Henshaw, Rensselaer Polytechnic Institute, USA Scalable Advection Algorithms for Multi-Tracers in Climate Codes Iulian Grindeanu, Argonne National Laboratory, USA; Kara Peterson, Sandia National Laboratories, USA; Vijay Mahadevan, Navamita Ray, and Rajeev Jain, Argonne National Laboratory, USA Fast Ship Hydrodynamics Via Novel Methods Stavros Kontos, Technical University of Denmark, Denmark; Ole Lindberg, FORCE Technology, Denmark; Allan Engsig-Karup and Harry Bingham, Technical University of Denmark, Denmark A Conservative, Positivity Preserving Scheme for Reactive Solute Transport Problems in Moving Domains Sibusiso Mabuza, University of Houston, USA Singly-Periodic Stokes Flow with a Wall Forest O. Mannan and Ricardo Cortez, Tulane University, USA Computational Hydrodynamics: How Portable and Scalable Are Heterogeneous Programming Paradigms? Wojciech Pawlak and Allan Engsig-Karup, Technical University of Denmark, Denmark; Stefan Glimberg, Lloyd’s Register Consulting, United Kingdom A Fully Discrete Derivation of a Direct Ale Conservative Scheme for Compressible Hydrodynamics Thibaud Vazquez-Gonzalez, Antoine Llor, and Christophe Fochesato, CEA, France
Sunday, March 15
PP2 Numerical PDEs 4:30 PM-6:30 PM Room:255 A Non Standard Scheme for Nagumo Type Differential Equations Adebayo A. Aderogba, Michael Chapwanya, and Pius Chin, University of Pretoria, South Africa Reduced Basis Methods for Calibration and Option Pricing Olena Burkovska, Kathrin Glau, Mirco Mahlstedt, and Barbara Wohlmuth, Technical University of Munich, Germany A Fast and Stable Explicit Operator Splitting Method for Phase-Field Models Yuanzhen Cheng, Alexander Kurganov, and Zhuolin Qu, Tulane University, USA; Tao Tang, Hong Kong Baptist University, Hong Kong Method of Lines Transpose Schemes for Parabolic Problems Hana Cho and Andrew J. Christlieb, Michigan State University, USA; Matt Causley, Kattering University, USA; David Seal, Michigan State University, USA Comparison of Nonlinear and Linear Stabilization Schemes for AdvectionDiffusion Equations Ryan R. Grove and Timo Heister, Clemson University, USA A Task-Parallel Approach for Solving PDEs on a Lattice John T. Hutchins, Derek A. Cline, and James C. Sutherland, University of Utah, USA Finite Element Analysis of Free Material Optimization Problems Tobias Jordan and Michael Hinze, University of Hamburg, Germany
continued in next column
81
Higher Order Numerical Schemes for Convection Diffusion Equation Based on B-Spline Quasi-Interpolation Rakesh Kumar and Sambandam Baskar, Indian Institute of Technology-Bombay, India NIST AMR Benchmarks William F. Mitchell, National Institute of Standards and Technology, USA Finite Element Methods for the Evolution Problem in General Relativity Vincent Quenneville-Belair, University of Minnesota, USA Massively Parallel Radiation Transport Sweeps on Unstructured Grids Jean C. Ragusa, Tarek Ghaddar, and Michael Adams, Texas A&M University, USA The Discrete Maximum Principle in the Family of Mimetic Finite Difference Methods Daniil Svyatskiy, Gianmarco Manzini, and Konstantin Lipnikov, Los Alamos National Laboratory, USA A Locally Adaptive RBF-FD Method Wade Meyers, University of Wisconsin, Stout, USA; Talin Masihimirzakhanian, California Polytechnic State University, Pomona, USA; Keith Wojciechowski, University of Wisconsin, Stout, USA A Second-Order Maximum Principle Preserving Lagrange Finite Element Technique for Nonlinear Scalar Conservation Equations Jean-Luc Guermond, Murtazo Nazarov, Bojan Popov, and Yong Yang, Texas A&M University, USA
82
Sunday, March 15
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
PP3
PP4
Linear Algebra and Data Analysis 4:30 PM-6:30 PM
Biomedical Applications 4:30 PM-6:30 PM Room:255
Room:255
Reduced Order Modelling for Optimal Cancer Treatment Bahodir Ahmedov, Aachen Institute for Advanced Study in Computational Engineering Science, Germany; Michael Herty and Martin Grepl, RWTH Aachen University, Germany
A New Test for Exclusion Algorithm to Find the Optimum Value of Function in Rn Ibraheem Alolyan, King Saud University, Saudia Arabia Approximation and Error Estimation in High Dimensional Space for Stochastic Collocation Methods on Arbitrary Sparse Samples Richard Archibald, Oak Ridge National Laboratory, USA Rational Least Squares Fitting using Krylov Spaces Mario Berljafa and Stefan Guettel, University of Manchester, United Kingdom A Hybrid Openmp/mpi Cg Iterative Eigensolver for First-Principles Plane Wave Materials Science Codes Andrew M. Canning, Lawrence Berkeley National Laboratory, USA All Real Eigenvalues of Symmetric Tensors Cui Chunfeng and Dai Yu-Hong, Chinese Academy of Sciences, China; Nie Jiawang, University of California, San Diego, USA A Python Toolbox for Shape Optimization in Imaging and Data Analysis Gunay Dogan, National Institute of Standards and Technology, USA Discovering Block Structure in Graphs with Approximate Eigenvectors James Fairbanks, Georgia Institute of Technology, USA; Geoffrey D. Sanders, Lawrence Livermore National Laboratory, USA Efficient Multigrid Methods for Distributed Optimal Control Problems Constrained by Parabolic Equations Mona Hajghassem and Andrei Draganescu, University of Maryland, Baltimore County, USA Computing the Heat Kernel of a Graph for a Local Clustering Algorithm Olivia Simpson, University of California, San Diego, USA
A Mesh Free Method for Numerical Simulation of Calcium Dynamics In Ventricular Myocytes Emmanuel O. Asante-Asamani, Bruce Wade, and Zeyun Yu, University of Wisconsin, Milwaukee, USA Computed Tear Film and Solute Dynamics on An Eye-Shaped Domain Richard Braun, University of Delaware, USA; Longfei Li, Rensselaer Polytechnic Institute, USA; Tobin Driscoll, University of Delaware, USA; William Henshaw and Jeffrey Banks, Rensselaer Polytechnic Institute, USA; P. Ewen King-Smith, The Ohio State University, USA The Transcriptomic Clock of Human Cerebral Cortex Development Elisabeth M. Brown, Kristin Bennett, Hannah De Los Santos, and Joey Lea, Rensselaer Polytechnic Institute, USA; Nathan Boles, Thomas Kiehl, Sally Temple, and Christopher Fasano, Neural Stem Cell Institute, USA Computational Methods to Study the Coordination of Mechanical Forces Involved in Amoeboid Cell Migration Calina A. Copos and Robert D. Guy, University of California, Davis, USA Segmentation and Processing of Brain Images of Multiple Modalities in 2 and 3 Dimensions John Edwards, Brian Summa, Valerio Pascucci, and Christopher Johnson, University of Utah, USA Improving Performance of MultiLevel Nonrigid Registration of Two Ct-Based Lung Images With Use of Gpu Computing Nathan Ellingwood and Youbing Yin, University of Iowa, USA; Matthew Smith, National Cheng Kung University, Taiwan; Ching-Long Lin, University of Iowa, USA continued in next column
Reaction of a Solid Tumor According to the Injection of Medical Supplies into Heart and Liver Jaegwi Go, Changwon National University, Korea Newtonian and Non-Newtonian Fluid Dynamics in Abdominal Aortic Aneurysms Danielle D. Masse and Jason Howell, College of Charleston, USA Topology Backs Alternative Medicine Claim Fernando Schwartz, University of Tennessee, USA; Louis Xiang, Chinese University of Hong Kong, Hong Kong; Kwai L. Wong, University of Tennessee and Oak Ridge National Laboratory, USA Shear Wave Filtering in Bouligand Structures Nicolas Guarin Zapata, Purdue University, USA; Juan Gomez, Universidad EAFIT, Colombia; Nick Yaraghi and David Kisailus, University of California, USA; Pablo Zavattieri, Purdue University, USA
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
PP5
Statistical Methods and Uncertainty Quantification 4:30 PM-6:30 PM Room:255 Adaptive Spectral Tensor-Train Decomposition for the Construction of Surrogate Models Daniele Bigoni and Allan Engsig-Karup, Technical University of Denmark, Denmark; Youssef M. Marzouk, Massachusetts Institute of Technology, USA A Posteriori Error Estimation for a Cut Cell Finite Volume Method in the Presence of Uncertainty James B. Collins, West Texas A&M University, USA; Simon Tavener and Don Estep, Colorado State University, USA Probability Measures on Numerical Solutions of Odes for Uncertainty Quantification and Inference Patrick R. Conrad, Massachusetts Institute of Technology, USA; Mark Girolami, University of Warwick, United Kingdom; Simo Sarkka, Aalto University, Finland; Andrew Stuart, University of Warwick, United Kingdom; Konstantinos Zygalakis, University of Southampton, United Kingdom Adaptive Bayesian Selection, Calibration, and Validation of Coarse-Grained Models of Atomistic Systems Kathryn Farrell, J. Tinsley Oden, and Danial Faghihi, University of Texas at Austin, USA Emgr - Empirical Gramian Framework Christian Himpe, University of Münster, Germany; Mario Ohlberger, University of Muenster, Germany Kernel Density Estimation for Implicit Monte Carlo Radiation Transport Aaron M. Holgado, Texas A&M University, USA; Robert Holladay, Virginia Tech, USA; Allan Wollaber, Mathew Cleveland, and Todd Urbatsch, Los Alamos National Laboratory, USA; Ryan McClarren, Texas A&M University, USA
continued in next column
83
Distribution Functions of Water Saturation for the Stochastic BuckleyLeverett Problem Via the Streamline Method Fayadhoi Ibrahima, Stanford University, USA; Daniel W. Meyer, Institute of Fluid Dynamics, Switzerland; Hamdi Tchelepi, Stanford University, USA
Sunday, March 15
Hybridized Reduced Basis Method and Generalized Polynomial Chaos for Solving Partial Differential Equations Jiahua Jiang, University of Massachusetts, Dartmouth, USA
Reproducible Numerical Computing with HashDist Aron Ahmadia, U.S. Army Engineer Research and Development Center, USA; Ondrej Certik, University of Nevada, Reno, USA; Christopher Kees, U.S. Army Engineer Research and Development Center, USA; Dag Sverre Seljebotn, University of Oslo, Norway; Andy R. Terrel, Continuum Analytics, USA
Kriging and Spatial Design Accelerated by Orders of Magnitude: Combining Low-Rank Covariance Approximations with FFT-Techniques Alexander Litvinenko, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Wolfgang Nowak, University of Stuttgart, Germany A Nonlinear Non-Gaussian Smoother for Continuous Stochastic Dynamical Systems Tapovan Lolla and Pierre Lermusiaux, Massachusetts Institute of Technology, USA Uncertainty Quantification in Incompressible Flow Using Sparse Grids Friedrich Menhorn and Tobias Neckel, Technische Universität München, Germany Matrix Splitting Techniques for Sampling a High-Dimensional Gaussian Richard A. Norton and Colin Fox, University of Otago, New Zealand Time Series Estimation of a Stochastic Processes Coupled to Pdes for Multiscale Modeling Charlie Vollmer and Don Estep, Colorado State University, USA; Anter A. El-Azab, Purdue University, USA Fast Stochastic Simulation of NonGaussian Correlated Process Variations Tsui-Wei Weng, Zheng Zhang, and Luca Daniel, Massachusetts Institute of Technology, USA Uncertainty Quantification for Integrated Circuits and {MEMS} Zheng Zhang, Massachusetts Institute of Technology, USA; Ibrahim Elfadel, Masdar Institute of Science and Engineering, United Arab Emirates; Luca Daniel, Massachusetts Institute of Technology, USA
PP6 CSE Methods 4:30 PM-6:30 PM Room:255
A Scalable Fast Method for $N$-Body Problems Based on Exact Finite Element Basis Screen Functions Natalie N. Beams, Luke Olson, and Jonathan B. Freund, University of Illinois at Urbana-Champaign, USA Boltzmann Equation Solver Adapted to Emergent Chemical NonEquilibrium Jeremiah Birrell, University of Arizona, USA; Jon Wilkening, Lawrence Berkeley National Laboratory and University of California Berkeley, USA; Cheng Tao Yang and Johann Rafelski, University of Arizona, USA Boltzmann Collision Operator for Cylindrically Symmetric Velocity Distributions in Plasmas Yanping Chen, Yannan Shen, John Zweck, and Matthew Goeckner, University of Texas at Dallas, USA Analysis of a Heterogeneous Multiscale Method for Poroelasticity Paul M. Delgado, Vinod Kumar, and Son Young Yi, University of Texas at El Paso, USA Cell List Algorithms for Nonequilibrium Molecular Dynamics Matthew Dobson, University of Massachusetts, Amherst, USA Convex-Hull Classification of Molecular Data on a Cluster Sally R. Ellingson and Radha Nagarajan, University of Kentucky, USA Direct Evaluation of Unified Extended Splines Ian D. Henriksen, Brigham Young University, USA continued on next page
84
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
PP6 CSE Methods 4:30 PM-6:30 PM continued
Spectral Noise Filtering for Fourier Transform Profilometry Thomas Höft, University of St. Thomas, USA Numerical Modeling of Wave Propagation in Poroelastic Media Using Optimal Staggered Implicit Finite Differences Ursula Iturraran-Viveros and Reymundo Itza, Universidad Nacional Autonoma de Mexico, Mexico Graph-Based Analysis of ThreeDimensional, Large Scale Phase-Field Simulations Marcus Jainta, Karlsruhe Institute of Technology, Germany; Daniel Stubenvoll, Johannes Hötzer, and Phillip Steinmetz, Karlsruhe University of Applied Sciences, Germany; Britta Nestler, Karlsruhe Institute of Technology, Germany Uncertainty Quantification for the Estimation of the Diffusion Coefficient from Md Simulations Changho Kim and George E. Karniadakis, Brown University, USA Adaptive Model Order Reduction in Forward and Inverse Multi-Frequency Problem for Maxwell’s Equations Michal A. Kordy, Elena Cherkaev, and Phil Wannamaker, University of Utah, USA Sparse Spectral Tau-Method for Binary Neutron Stars Stephen Lau, University of New Mexico, USA Time-Parallel Approaches for Complex Rotorcraft Calculations Joshua I. Leffell, NASA Ames Research Center, USA; Jay Sitaraman, University of Wyoming, USA; Andrew Wissink, NASA Ames Research Center, USA Precice -- Flexible Parallel MultiPhysics Coupling Florian Lindner and Miriam Mehl,
continued in next column
Universität Stuttgart, Germany; Benjamin Uekermann, Technische Universität München, Germany Matched Asymptotic Analysis to Solve the Narrow Escape Problem in a Domain with a Long Neck Xiaofei Li and Hyundae Lee, Inha University, Korea Computational Homogenization for the Modeling of Soft Matter Materials Christian Linder, Stanford University, USA A Sparse Interpolation Algorithm for Dynamical Simulations in Computational Chemistry James Nance, ELENA Jakubikova, and C.T. Kelley, North Carolina State University, USA P²NFFT - A Versatile Framework for Computing NFFT-based Fast Ewald Summation Michael Pippig and Franziska Nestler, Chemnitz University of Technology, Germany
Sunday, March 15
PP7
CSE Applications 4:30 PM-6:30 PM Room:255 Fields That Cause Elastic Breakdown in Inhomogeneous Media Nathan C. Briggs, Graeme Milton, Zoe Last Koch, Andrew Boyles, Jonathan Boyle, Michael Primrose, and Michael Zhao, University of Utah, USA Charge Transfer Processes at Semiconductor-Electrolyte Interfaces in Solar Cell Modeling Michael D. Harmon, Institute for Computational Engineering and Sciences, USA Student Chapter Develops Future Professionals Stacey Joseph-Ellison, Embry-Riddle Aeronautical University, USA Optimal Control of Miscible Displacement Equations Using Discontinuous Galerkin Methods Brianna Lynn, Rice University, USA Identifying and Tracking Multiple Underwater Acoustic Sources Using Characteristic Signatures Zoi-Heleni Michalopoulou, Jacob Moorman, Jake Brusca, and Shan Fung, New Jersey Institute of Technology, USA Elastic Deformation Due to Dislocations in a Transversely Isotropic Viscoelastic Halfspace Amirhossein Molavi Tabrizi, Ernian Pan, and Ali Sangghaleh, University of Akron, USA Simulation and Modeling of Unmanned Systems for Humanitarian Applications in Industry Courtney E. Thurston, Commonwealth Connections Academy, USA
2015 SIAM Conference on Computational Science and Engineering
85
Sunday, March 15
Sunday, March 15
Sunday, March 15
PP8
PP101
PP102
AWM Workshop 4:30 PM-6:30 PM
Minisymposterium -BET: Open Source Software for Stochastic Inverse Problems in a MeasureTheoretic Context 4:30 PM-6:30 PM
Minisymposterium -Clawpack Development, Extensions and Applications 4:30 PM-6:30 PM
Room:255 AWM Workshop: Sampling and Reconstruction in Inite-Dimensional Reproducing Kernel Subspace Cheng Cheng, University of Central Florida, USA AWM Workshop: A Lattice of Poincare Duality Algebras with Acyclic Annihilators and Finite Dimension Associated to a Manifold Cameron Crowe, Stony Brook University, USA
Room:255 Organizer: Lindley C. Graham, University of Texas at Austin, USA Bet: Algorithmic and Error Analyses Troy Butler, University of Colorado, Denver, USA
AWM Workshop: Residual Based Aposteriori Error Estimation in a Fully Automatic Hp –fem for the Stokes Equations Arezou Ghesmati, Markus Buerg, and Wolfgang Bangerth, Texas A&M University, USA
BET: Applications for an Open Source Inverse Problems Package Lindley C. Graham and Steven Mattis, University of Texas at Austin, USA; Troy Butler, University of Colorado, Denver, USA; Clint Dawson, University of Texas at Austin, USA
AWM Workshop: Residual-Based A Posteriori Error Estimate for Interface Problems: Nonconforming Linear Elements Cuiyu He and Zhiqiang Cai, Purdue University, USA; Shun Zhang, City University of Hong Kong, Hong Kong
Bet: Modifications and Analysis for Model Discrepancies Nishant Panda, Colorado State University, USA
AWM Workshop: Enhancements for Reduced Basis Methods: Reducing Offline Computational Costs Jiahua Jiang, University of Massachusetts, Dartmouth, USA AWM Workshop: Combinatorial Navier-Stokes Equation Aradhana Kumari, City University of New York, USA AWM Workshop: An Adaptive Gmsfem for High-Contrast Flow Problems Guanglian Li, Texas A&M University, USA AWM Workshop: Propagation Failure in Discrete Inhomogeneous Media Using a Caricature of the Cubic Elizabeth Lydon, University of Central Florida, USA AWM Workshop: Nontrivial Structure in Top Homology of a Space Chandrika Sadanand, Stony Brook University, USA
Room:255 Organizer: Randall LeVeque, University of Washington, USA Tsunami Modeling In North Africa Using Geoclaw Software: a Tool for the Tsunami Scenario Database in the West Mediterranean Lubna Amir, USTHB University, Algeria; Walter Dudley, University of Hawaii at Hilo, USA; Jean Roger, G-Mer Etudes Marines, France ForestClaw : Parallel, Adaptive, Multiblock Simulations for Clawpack Donna Calhoun, Boise State University, USA Adjoint Methods for Guiding Adaptive Mesh Refinement in Wave Propagation Problems Brisa Davis and Randall LeVeque, University of Washington, USA A Community-Driven Collection of Approximate Riemann Solvers for Hyperbolic Problems Mauricio J. Del Razo, University of Washington, USA; David I. Ketcheson, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Randall LeVeque, University of Washington, USA High Resolution Tsunami Modeling at the Mediterranean Coast of Israel Towards An Early Warning Tsunami Scenarios Data Bank Barak Galanti, Israel Oceanographic and Limnological Research, Israel PyClaw: Accurate, Scalable Solution of Hyperbolic PDEs in Python David I. Ketcheson, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Aron Ahmadia, U.S. Army Engineer Research and Development Center, USA; Kyle T. Mandli, Columbia University, USA
continued on next page
86
Sunday, March 15
PP102 Minisymposterium -Clawpack Development, Extensions and Applications 4:30 PM-6:30 PM continued
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
PP103 Minisymposterium -Frameworks, Algorithms, and Scalable Technologies for Mathematics (FASTMath) 4:30 PM-6:30 PM Room:255
Practical Applications of GeoClaw to Tsunami Hazard Assessment Randall LeVeque, Loyce Adams, and Frank I. Gonzalez, University of Washington, USA CUDACLAW: A GPU Framework for the Solution of Hyperbolic Pdes George M. Turkiyyah, American University of Beirut, Lebanon; H. Gorune Ohannessian, University of Wisconsin, USA; Aron Ahmadia, U.S. Army Engineer Research and Development Center, USA; David I. Ketcheson, King Abdullah University of Science & Technology (KAUST), Saudi Arabia
Organizer: Lori A. Diachin, Lawrence Livermore National Laboratory, USA An Overview of PETSc Satish Balay, Argonne National Laboratory, USA; Jed Brown, Argonne National Laboratory, USA and University of Colorado Boulder, USA; William D. Gropp, University of Illinois at UrbanaChampaign, USA; Matthew Knepley, University of Chicago, USA; Lois Curfman McInnes, Barry F. Smith, and Hong Zhang, Argonne National Laboratory, USA Construction of Parallel Adaptive Simulation Loops Brian Granzow, Rensselaer Polytechnic Institute, USA MueLu: Multigrid Framework for Advanced Architectures Jonathan J. Hu and Andrey Prokopenko, Sandia National Laboratories, USA Parallel Unstructured Mesh Infrastructure Dan A. Ibanez, E. Seegyoung Seol, and Gerrett Diamond, Rensselaer Polytechnic Institute, USA Massively Parallel Adaptive Simulations Using Petsc for Turbulent Boundary Layer Flows Michel Rasquin, University of Colorado Boulder, USA; Dan A. Ibanez, Rensselaer Polytechnic Institute, USA; Benjamin Matthews, University of Colorado Boulder, USA; Cameron Smith, Onkar Sahni, and Mark S. Shephard, Rensselaer Polytechnic Institute, USA; Kenneth Jansen, University of Colorado Boulder, USA
continued in next column
Sparse Direct Solvers on Distributed CPU-GPU Machines (unconfirmed) Pieter Ghysels, Xiaoye Sherry Li, and Francois-Henry Rouet, Lawrence Berkeley National Laboratory, USA; Piyush Sao and Richard Vuduc, Georgia Institute of Technology, USA Dynamic Partitioning Using Mesh Adjacencies Cameron Smith, Dan A. Ibanez, and Gerrett Diamond, Rensselaer Polytechnic Institute, USA Fastmath Structured Mesh and Particle Technologies Anshu Dubey, Phillip Colella, Mark Adams, Ann S. Almgren, Dan Graves, Terry J. Ligocki, and Brian Van Straalen, Lawrence Berkeley National Laboratory, USA; Milo Dorr, Lawrence Livermore National Laboratory, USA Sundials: Suite of Nonlinear and Differential/algebraic Solvers Daniel R. Reynolds, Southern Methodist University, USA; Carol S. Woodward, Lawrence Livermore National Laboratory, USA Hypre: High Performance Preconditioners Robert Falgout, Tzanio V. Kolev, Jacob B. Schroder, and Ulrike M. Yang, Lawrence Livermore National Laboratory, USA
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
Sunday, March 15
PP104
PP105
Minisymposterium -Integrated Mathematics for Mesoscopic Modeling of Materials 4:30 PM-6:30 PM
Minisymposterium -Numerical Methods for Kinetic Equations and Related Models 4:30 PM-6:30 PM
Room:255
Room:255
Organizers:
Organizer: Martin Frank, RWTH - Aachen University of Technology, Germany
George E. Karniadakis, Brown University, USA Nathan Baker, Pacific Northwest National Laboratory, USA Stochastic Methods in Mesoscopic Materials Modeling Paul J. Atzberger, University of California, Santa Barbara, USA Concurrent Coupling Methods in Mesoscopic Materials Modeling Xin Bian, Brown University, USA Overview of Mathematics for Mesoscopic Modeling of Materials George E. Karniadakis, Brown University, USA Hierarchical Coarse-graining and Parallelization Methods for Mesoscale material Models Markos A. Katsoulakis, University of Massachusetts, Amherst, USA Particle-Based Methods in Mesoscopic Materials Modeling Zhen Li, Brown University, USA Grid-Based Methods in Mesoscopic Materials Modeling Martin Maxey, Brown University, USA Applications in Mesoscopic Modeling of Materials Wenxiao Pan, Pacific Northwest National Laboratory, USA; Mauro Perego, Sandia National Laboratories, USA Coarse-Graining in Mesoscopic Materials Modeling Panos Stinis, University of Minnesota, USA Ultra Coarse-Graining in Mesoscopic Materials Modeling Gregory Voth, James Dama, and John Grime, University of Chicago, USA Fast Solvers for Mesoscopic Materials Modeling Jinchao Xu and Chun Liu, Pennsylvania State University, USA; Xiaozhe Hu, Tufts University, USA; Maximilian S. Metti, Pennsylvania State University, USA
Realizability Limiting in High-Order Numerical Solutions of Entropy-Based Moment Closures Graham Alldredge, RWTH Aachen University, Germany; Florian Schneider, Technische Universität Kaiserslautern, Germany Exploration and Validation of FullDomain Massively Parallel Transport Sweep Algorithms Teresa S. Bailey, Peter Brown, and Adam Kunen, Lawrence Livermore National Laboratory, USA A New Moment Method in the Kinetic Theory of Gases Based on the L2 Function Space Zhenning Cai and Manuel Torrilhon, RWTH Aachen University, Germany Markov Chain Formalism for Radiative Transfer in Planetary Atmospheres: Forward Modeling, Including Linearization Anthony B. Davis, Feng Xu, Robert West, and David Diner, California Institute of Technology, USA On the Hyperbolicity of Grad’s 13 Moment System Yuwei Fan, Peking University, China; Zhenning Cai, RWTH Aachen University, Germany; Ruo Li, Peking University, China Convergence of Filtered Spherical Harmonic Equations for Radiation Transport Martin Frank, RWTH - Aachen University of Technology, Germany; Cory Hauck, Oak Ridge National Laboratory, USA; Kerstin Kuepper, RWTH - Aachen University of Technology, Germany
continued in next column
87
A High Order, Implicit, Hybrid Solver for Linear Kinetic Equations Michael Crockat, Michigan State University, USA; Charles K. Garrett and Cory Hauck, Oak Ridge National Laboratory, USA A High Order Time Splitting Method Based on Integral Deferred Correction for Semi-Lagrangian Vlasov Simulations Wei Guo, Michigan State University, USA Analysis of Discontinuous Galerkin Algorithms for Diffusion and for EnergyConserving Hamiltonian Dynamics Greg Hammett, Princeton University, USA; Ammar Hakim, Princeton Plasma Physics Laboratory, USA; Eric Shi, Ian Abel, and Tim Stoltzfus-Dueck, Princeton University, USA Massively Parallel Calculations of Neutronics Experiments Using Pdt Marvin L. Adams, Aaron Holzaepfel, W. Daryl Hawkins, Michael Adams, Anthony Barbu, and Timmie Smith, Texas A&M University, USA Numerical Solution of the Boltzmann Equation Using Quadrature-Based Projection Methods Julian Koellermeier and Manuel Torrilhon, RWTH Aachen University, Germany Positive Filtered PN Closures for Linear Kinetic Transport Equations, with some Convergence Results Ming Tse P. Laiu, University of Maryland, College Park, USA; Cory Hauck, Oak Ridge National Laboratory, USA; Dianne P. O’Leary and André Tits, University of Maryland, College Park, USA Implicit, Filtered Pn Methods for Radiation Transport Ryan G. McClarren and Vincent Laboure, Texas A&M University, USA; Cory Hauck, Oak Ridge National Laboratory, USA Massively Parallel Nuclear Reactor Analysis Using Pdt Marvin L. Adams, Carolyn McGraw, W. Daryl Hawkins, Michael Adams, and Timmie Smith, Texas A&M University, USA High Order Asymptotic Preserving Nodal Discontinuous Galerkin Imex Schemes for the Bgk Equation Jingmei Qiu, University of Houston, USA; Juhi Jang, University of California, Riverside, USA; Fengyan Li, Rensselaer Polytechnic Institute, USA; Tao Xiong, University of Houston, USA continued on next page
88
Sunday, March 15
PP105
2015 SIAM Conference on Computational Science and Engineering
Sunday, March 15
PP106
Minisymposterium -Numerical Methods for Kinetic Equations and Related Models 4:30 PM-6:30 PM
Minisymposterium -Scalable Finite Element Assembly 4:30 PM-6:30 PM
continued
Organizers: Irina Demeshko, Sandia National Laboratories, USA
Towards Hyperbolic Moment Approximations of Multicomponent Plasmas Roman P. Schärer and Manuel Torrilhon, RWTH Aachen University, Germany
Eric C. Cyr, Sandia National Laboratories, USA
Ifp: An Optimal, Fully Conservative, Fully Implicit, Vlasov-Fokker-Planck Solver: Poster William T. Taitano, Los Alamos National Laboratory, USA Residual Monte Carlo Methods Within the Moment-Based Acceleration Framework Jeffrey A. Willert, Los Alamos National Laboratory, USA Asymptotic Preserving Discontinuous Galerkin Method for the Radiative Transfer Equation Yulong Xing, University of Tennessee and Oak Ridge National Laboratory, USA; Cory Hauck, Oak Ridge National Laboratory, USA
Room:255
The scalability of PDE codes typically focuses on solving large sparse linear system of equations resulting from a FEM discretization. Finite element assembly is often dismissed as inconsequential compared to the expense of linear solves. However, due to the increased parallelism available on next-generation architectures the scalability of the assembly can became a bottleneck to achieving performance. The aim of this minisymposterium is provide a forum to discuss ideas for developing a portable scalable finite element assembly on modern and next-generation architectures. Architecture Portable Assembly for Maxwell’s Equations Eric C. Cyr, Irina Demeshko, Roger Pawlowski, and Matthew Bettencourt, Sandia National Laboratories, USA Towards Exascale Implementation of the Finite Element Based Application Development Environment Irina Demeshko, H. Carter Edwards, Michael Heroux, ROGER P. Pawkowski, Eric Phipps, and Andrew Salinger, Sandia National Laboratories, USA Multicore Finite Element Assembly Via Scans Robert C. Kirby, Baylor University, USA Operator Transformation and Code Generation for Scientific Computing Andreas Kloeckner, University of Illinois, USA OCCA: A Unified Approach to MultiThreading Languages David Medina and Tim Warburton, Rice University, USA; Amik St-Cyr, Shell International Exploration & Production B.V., Netherlands
continued in next column
Assembly Algorithms for Pdes with Uncertain Input Data on Emerging Multicore Architectures Eric Phipps and H. Carter Edwards, Sandia National Laboratories, USA Using Multicore Parallelism for Common Finite Element Operations Bruno Turcksin, Texas A&M University, USA; Martin Kronbichler, Technische Universität München, Germany; Wolfgang Bangerth, Texas A&M University, USA
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
Registration 7:30 AM-5:00 PM
DAG-Based Efficient Scalable and Portable PDE Software 9:10 AM-10:50 AM
Room:East Foyer
Announcements 8:10 AM-8:15 AM Room:355
IP5 Statistical and Computational Challenges of Constraining Greenhouse Gas Budgets 8:15 AM-9:00 AM Room:355 Chair: Linda R. Petzold, University of California, Santa Barbara, USA Predicting future changes to the global carbon cycle (and therefore climate) and quantifying anthropogenic emissions of greenhouse gases (GHGs) both require an understanding of net GHGs emissions and uptake across a variety of spatial and temporal scales. This talk will explore some of the core scientific questions related to understanding GHG budgets through the lens of the statistical and computational challenges that arise. The focus will be on the use of atmospheric observations, and applications will include the natural and anthropogenic components of the methane and carbon dioxide budgets. The discussion will include issues related to the solution of spatiotemporal inverse problems, uncertainty quantification, data fusion, gap filling, and issues of “big data” arising from the use of satellite observations.
Anna Michalak Carnegie Institution for Science and Stanford University, USA
Intermission 9:00 AM-9:10 AM
MS129
Room:355 Directed acyclic graph (DAG) based approaches can be an effective way to manage complexity in large scale pde computations. Several different examples as well as an overview of existing approaches will be given. The examples shown will demonstrate the versatility of the approach and give some indications of its promise for present and future large scale computations arising from challenging multi-scale and multi-physics partial differential equations simulations on both multi-core and heterogeneous architectures.
Organizer: Martin Berzins University of Utah, USA 9:10-9:30 Using Multiple Dags to Ensure Portability and Scalability in Large Scale Computations Using Uintah John A. Schmidt, University of Utah, USA 9:35-9:55 A Comparative Analysis of of Asynchronous Many Task Programming Models for Next Generation Platforms Janine C. Bennett, H. Kolla, Jeremiah Wilke, Keita Teranishi, Nicole Slattengren, Greg Sjaardema, and Samuel Knight, Sandia National Laboratories, USA 10:00-10:20 Structured Dagger: Supporting Asynchrony with Clarity Jonathan Lifflander, University of Illinois, USA 10:25-10:45 A DAG Approach to Tame Complexity in Multiphysics Software on Heterogeneous Architectures James C. Sutherland and Abhishek Bagusetty, University of Utah, USA
89
Monday, March 16
MS130 Cut Cells: Algorithms and Applications - Part I of II 9:10 AM-10:50 AM Room:355 A For Part 2 see MS157 Cut cell methods solve PDEs in complex geometry using a regular Cartesian grid with special treatment for cells that intersect the boundary. These methods are attractive because they easily represent complex and moving geometries. Cut cell methods have been used with finite difference, volume and element discretizations, across a wide range of problems, including aerodynamics, meteorology, oceanography, and others. Challenges include gaps in theoretical underpinnings; issues due to arbitrarily small cells; and the programming complexity that arises with cell merging and other stabilization techniques. This minisymposium will highlight recent algorithmic advances, and include a variety of science domain applications.
Organizer: Hans Johansen Lawrence Berkeley National Laboratory, USA
Organizer: Marsha Berger Courant Institute of Mathematical Sciences, New York University, USA 9:10-9:30 Weno Finite Volume Methods for Embedded Boundary Grids Christiane Helzel, Ruhr-Universität Bochum, Germany 9:35-9:55 High-Order Quadrature on Implicitly Defined Domains with Application to a High-Order Embedded Boundary Discontinuous Galerkin Method for Evolving Interface Problems Robert Saye, Lawrence Berkeley National Laboratory, USA 10:00-10:20 Terrain Following Versus Cut-Cells Hilary Weller and James Shaw, University of Reading, United Kingdom 10:25-10:45 Diffusion MRI on a Cartesian Grid with Immersed Interfaces Khieu Van Nguyen, CEA Saclay, France; Jing-Rebecca Li, INRIA Saclay and CMAP Ecole Polytechnique, France; Luisa Ciobanu, CEA Saclay, France
90
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
Monday, March 16
MS131
MS132
Report on the Future of CSE Education and Research 9:10 AM-10:50 AM
Recent Advances in Bayesian Methods for Computational Science and Engineering - Part I of II 9:10 AM-10:50 AM
High-dimensional Approximation and Integration: Analysis and Computation - Part III of V 9:10 AM-10:50 AM
Room:250 A
Room:250 B
For Part 2 see MS158 This minisymposium highlights the most recent impact of Bayesian inference for uncertainty quantification, statistical modeling, experimental design, and optimization. We discuss novel approximation techniques bringing with them a myriad of new potential applications from medicine to engineering, finance, and biology. The minisymposium brings together researchers from diverse application domains to introduce and transfer recent contributions, as well as to discuss open challenges due to increasing demand for efficient techniques.
University of California, Santa Barbara, USA
For Part 2 see MS106 For Part 4 see MS159 This minisymposium focuses on the fundamental problem of how to accurately approximate solutions of both forward and inverse high-dimensional functions. Predicting the behavior of complex phenomena relies on constructing solutions in terms of high dimensional spaces, particularly in the case when the input data (coefficients, forcing terms, initial and boundary conditions, geometry) are affected by large amounts of uncertainty. The resulting explosion in computational effort is a symptom of the curse of dimensionality and this symposium aims at exploring breakthroughs in sparse sampling and representations, nonlinear and greedy approximations, compressed sensing, multilevel methods, and ”best-N-term” approximations.
Organizer: James Beck
Organizer: Clayton G. Webster
California Institute of Technology, USA
Oak Ridge National Laboratory, USA
9:10-9:30 Bayesian Updating for Dynamic Systems Using Subset Simulation (Beck) and Active Model Selection (Busetto) AlbertoGiovanni Busetto, University of California, Santa Barbara, USA; James Beck, California Institute of Technology, USA
Organizer: Guannan Zhang
Room:355 D This session will present and discuss a draft white paper on “Future Directions in CSE Education and Research” which was composed in 2014 by a subset of the CSE community under the guidance of the Officers of the SIAM Activity Group on CSE. A major goal of this session is to solicit feedback on this document from the broad CSE community. The new document follows in the footsteps of the 2001 report on “Graduate Education in CSE” (Petzold et al.). The white paper describes and investigates the rapid expansion of CSE since the beginning of the 21st century and the challenges the CSE field is encountering in the context of recent disruptive developments that include extreme-scale computing, data-driven discovery, and a comprehensive broadening of the application fields of CSE. There will be extensive opportunity for feedback and discussion, and these challenges and opportunities will also be discussed further in today’s lunchtime Forward Looking Panel.
Organizer: Hans De Sterck University of Waterloo, Canada
Organizer: Ulrich J. Ruede University of Erlangen-Nuremberg, Germany
Organizer: Lois Curfman McInnes Argonne National Laboratory, USA
Organizer: Karen E. Willcox Massachusetts Institute of Technology, USA 9:10-9:30 The Future of CSE Research Ulrich J. Ruede, University of ErlangenNuremberg, Germany; Lois Curfman McInnes, Argonne National Laboratory, USA
Organizer: AlbertoGiovanni Busetto
9:35-9:55 Bayesian Inference of Chemical Kinetic Models from Proposed Reactions Nikhil Galagali and Youssef M. Marzouk, Massachusetts Institute of Technology, USA 10:00-10:20 Advanced Bayesian Computation for Challenging Problems in the Sciences and Engineering Mark Girolami, University of Warwick, United Kingdom
MS133
Oak Ridge National Laboratory, USA
Organizer: Max Gunzburger Florida State University, USA
Organizer: Albert Cohen Université Pierre et Marie Curie, France 9:10-9:30 Multilevel Simulation of Mean Exit Times Michael B. Giles, University of Oxford, United Kingdom; Francisco Bernal, Instituto Superior Tecnico, Portugal 9:35-9:55 A Multilevel Stochastic Collocation Method for Pdes with Random Inputs Aretha Teckentrup, University of Bath, United Kingdom; Peter Jantsch, University of Tennessee, USA; Max Gunzburger, Florida State University, USA; Clayton G. Webster, Oak Ridge National Laboratory, USA
10:25-10:45 Topological Sensitivity Analysis in Systems Biology Ann C. Babtie, Paul Kirk, and Michael Stumpf, Imperial College London, United Kingdom continued on next page
2015 SIAM Conference on Computational Science and Engineering
10:00-10:20 Estimating the N-Width of Solution Manifolds of Parametric Pdes Albert Cohen, Université Pierre et Marie Curie, France 10:25-10:45 Hierarchical Acceleration of Stochastic Collocation Methods for PDEs with Random Input Data Guannan Zhang and Clayton G. Webster, Oak Ridge National Laboratory, USA
91
Monday, March 16
Monday, March 16
MS134
MS135
Resilience in Numerical Simulations and Algorithms at Extreme Scale: Part III of IV - Sparse Matrix Methods 9:10 AM-10:50 AM
Model Error Assessment in Computational Physical Models - Part I of II 9:10 AM-10:50 AM
Room:250 C
For Part 2 see MS161 While parametric uncertainty quantification is performed routinely in computational science and engineering, model error assessment is still largely lacking and faces significant challenges that are yet to be resolved. For example, it is unclear how to properly disambiguate data noise from model errors. Recently there has been growing interest and some development towards the quantification of model errors, i.e. the inherent discrepancies from the truth that cannot be reduced within the model assumptions. This minisymposium focuses on state-of-the-art and novel methods for quantification of model errors. It highlights challenges and introduces a range of available methods to the computational science community, hoping to contribute significantly to improving predictive fidelity of computational physical models across a range of disciplines.
For Part 2 see MS107 For Part 4 see MS160 The advent of extreme scale machines will require the use of parallel resources at an unprecedented scale, probably leading to a high rate of hardware faults. Handling fully these faults at the computer system level may have a prohibitive cost. High performance computing applications that aim at exploiting all these resources will thus need to be resilient, in this minisympoisum different complementary approaches and methods will be presented to possibly address this key aspect of extreme scale computing.
Organizer: Keita Teranishi Sandia National Laboratories, USA
Organizer: Luc Giraud INRIA, France
Organizer: Michael Heroux Sandia National Laboratories, USA
Organizer: Emmanuel Agullo INRIA, France 9:10-9:30 Inherent Error Resilience of a Complex Moment-Based Eigensolver Akira Imakura, Yasunori Futamura, and Tetsuya Sakurai, University of Tsukuba, Japan 9:35-9:55 On the Reliability of Soft Error Detection in CGPOP Agullo Emmanuel, Luc Giraud, and Emrullah Fatih Yetkin, INRIA, France 10:00-10:20 Analysis of Krylov Solver Resilience in the Presence of SoftFaults Miroslav Stoyanov and Clayton G. Webster, Oak Ridge National Laboratory, USA 10:25-10:45 The Cost of Reliability: Iterative Linear Solvers and Reactive Fault Tolerance James Elliott, North Carolina State University, USA; Mark Hoemmen, Sandia National Laboratories, USA; Frank Mueller, North Carolina State University, USA
Room:250 D
Organizer: Khachik Sargsyan Sandia National Laboratories, USA
Organizer: Habib N. Najm Sandia National Laboratories, USA 9:10-9:30 Accounting for Model Error in the Calibration of Physical Models Habib N. Najm and Khachik Sargsyan, Sandia National Laboratories, USA; Roger Ghanem, University of Southern California, USA 9:35-9:55 Predictive Rans Simulations Via Bayesian Model-Scenario Averaging Richard Dwight, Delft University of Technology, Netherlands; Wouter Edeling, TU Delft, Netherlands; Paola Cinnella, ENSAM, ParisTech, France
continued on next page
92
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
MS135
MS136
MS137
Model Error Assessment in Computational Physical Models - Part I of II 9:10 AM-10:50 AM
Advances in Time-stepping Methods - Part I of V 9:10 AM-10:50 AM Room:250 E
continued
For Part 2 see MS162 Time-stepping methods are numerical methods for the time evolution of ODEs, DAEs, and PDEs in the method-of-lines framework. Many applications require specialized time-stepping methods in order to enhance efficiency or capture particular theoretical properties such as positivity, symplecticity, and strongstability-preservation or practical properties such as parallelization on emerging highperformance architectures, component (e.g., multirate or local time-stepping) and physics (IMEX or semi-implicit) splittings based on sub-system properties, and reliable error estimation and control. The talks in this minisymposium will describe recent novel developments in time-stepping methods and demonstrate their benefits in practical applications.
Advanced Finite Element Methods for Nonlinear Materials and Fluids Part III of III 9:10 AM-10:50 AM
Monday, March 16
10:00-10:20 Model Calibration and Error Propagation for Large-Eddy Simulation of Turbulent Flows Myra Blaylock, Cosmin Safta, Stefan P. Domino, John C. Hewson, and Jeremy Templeton, Sandia National Laboratories, USA 10:25-10:45 Bayesian Model Calibration Techniques That Incorporate Mixed Effects and Model Discrepancy Brian Williams, Los Alamos National Laboratory, USA; Kathleen Schmidt and Ralph C. Smith, North Carolina State University, USA
Organizer: Emil M. Constantinescu Argonne National Laboratory, USA
Organizer: Michael Guenther Bergische Universität, Germany
Organizer: Adrian Sandu Virginia Tech, USA
Organizer: Raymond J. Spiteri University of Saskatchewan, Canada 9:10-9:30 Generalized Structure Additive Runge-Kutta Methods Adrian Sandu, Virginia Tech, USA; Michael Guenther, Bergische Universität, Germany 9:35-9:55 Resolvent Expansions for Higher-order Simulations of PDEs Andrew J. Christlieb, Michigan State University, USA; Matthew Causley, Kettering University, USA; Hana Cho, Michigan State University, USA 10:00-10:20 K-Methods, An Extension of Exponential and Rosenbrock Time Integrators Paul Tranquilli, Adrian Sandu, and Ross Glandon, Virginia Tech, USA 10:25-10:45 Efficient Exponential Integrators: Construction, Analysis and Implementation Mayya Tokman, University of California, Merced, USA
Room:250 F For Part 2 see MS110 Numerical modeling of materials and fluids with nonlinear behavior plays an increasing role in computational science and engineering, particularly in the highfidelity simulation of elastic and plastic deformations, magneto-sensitive materials, liquid crystals, biological tissues and blood flow, among others. Despite a long history of efforts, there is still a need for the development of advanced finite-element methods with specific properties such as accurate representation of stresses, forces, and other quantities depending on the application. In many of these applications, mesh-adaptive implementations and efficient solvers are an important issue due to the size and complexity of the problems.
Organizer: James H. Adler Tufts University, USA
Organizer: Gerhard Starke University of Duisburg-Essen, Germany
Organizer: Thomas Manteuffel University of Colorado Boulder, USA 9:10-9:30 Nested Iteration and First-Order System Least Squares for Preconditioning a Two-Fluid Electromagnetic Plasma Model Chris Leibs and Thomas Manteuffel, University of Colorado Boulder, USA 9:35-9:55 An Energy-Minimization Finite-Element Approach for the Frank-Oseen Model of Nematic Liquid Crystals David B. Emerson, James H. Adler, Scott Maclachlan, and Timothy Atherton, Tufts University, USA
continued on next page
2015 SIAM Conference on Computational Science and Engineering
93
10:00-10:20 Energy Laws and FirstOrder System Least Squares for MHD systems Ilya Lashuk, Georgia Institute of Technology, USA; James H. Adler and Scott Maclachlan, Tufts University, USA; Ludmil Zikatanov, Pennsylvania State University, USA
Monday, March 16
Monday, March 16
MS138
MS139
UQ in Large Scale Computing - Part I of II 9:10 AM-10:50 AM
10:25-10:45 Quantity-of-Interest Based Least-Squares Finite Element Methods Jehanzeb H. Chaudhry, Florida State University, USA; Thomas Manteuffel, University of Colorado Boulder, USA; Luke Olson, University of Illinois at Urbana-Champaign, USA; Eric C. Cyr, Sandia National Laboratories, USA; Kuo Liu and Lei Tang, University of Colorado Boulder, USA
Room:251 A
Inverse Analysis and Uncertainty Quantification in Fluid Mechanics Part I of II 9:10 AM-10:50 AM
For Part 2 see MS164 The move towards ever-larger scale computing is bringing both challenges and opportunities to the field of uncertainty quantification. On the one hand, the increase in computing power helps to offset the cost of UQ, even for complex applications. On the other hand, communication bottlenecks, increased frequencies of soft and hard faults, and limits on memory create computational challenges for UQ algorithms. Furthermore, the relative sparsity of validation data for high-dimensional, complex applications puts strenuous demands on UQ to assess the predictive fidelity of the simulations. This minisymposium brings together talks that address these challenges.
Organizer: Bert J. Debusschere Sandia National Laboratories, USA
Organizer: Eric Phipps Sandia National Laboratories, USA
Room:251 B For Part 2 see MS165 Advances in computing resources and algorithms have enabled simulations of engineered systems of unprecedented complexity. Increasingly, these simulations are being used to inform important design and operational decisions. In this context, providing defensible uncertainty estimates for computed quantities of interest is critical. The aim of this minisymposium is to bring together researchers developing approaches for quantifying uncertainties, including those due to uncertain parameters, inadequate physical models, and uncertain or sparse experimental data, in fluid mechanics applications. There is broad interest in both statistical approaches and fluids mechanics applications for participation in this minisymposium.
9:10-9:30 Parametric Uncertainty Propagation in Resilient Domain Decomposition Methods Paul Mycek, Duke University, USA; Olivier P. Le Maitre, LIMSI-CNRS, France; Francesco Rizzi, Khachik Sargsyan, Karla Morris, and Bert J. Debusschere, Sandia National Laboratories, USA; Omar M. Knio, Duke University, USA
Organizer: Paul Bauman
9:35-9:55 Exploring Embedded Uncertainty Quantification Methods on Next-Generation Computer Architectures Eric Phipps, H. Carter Edwards, and Jonathan J. Hu, Sandia National Laboratories, USA
9:10-9:30 Comparison of Laminar Flame Models in the Presence of Uncertainty Paul Bauman, State University of New York, Buffalo, USA
10:00-10:20 Hierarchically Accelerated Stochastic Collocation for Random PDEs Peter Jantsch, University of Tennessee, USA; Diego Galindo, Clayton G. Webster, and Guannan Zhang, Oak Ridge National Laboratory, USA 10:25-10:45 A Multilevel Solution Strategy for the Stochastic Galerkin Method for PDEs with Random Coefficients Sarah Osborn and Victoria Howle, Texas Tech University, USA; Jonathan J. Hu and Eric Phipps, Sandia National Laboratories, USA
State University of New York, Buffalo, USA
Organizer: Todd Oliver University of Texas at Austin, USA
Organizer: David Salac University of Buffalo, SUNY, USA
Organizer: Abani K. Patra State University of New York, Buffalo, USA
9:35-9:55 Towards Experimental Design Strategies for Inadequate Models Gabriel Terejanu and Xiao Lin, University of South Carolina, USA 10:00-10:20 Liposome Vesicles in the Presence of Uncertainty David Salac, University of Buffalo, SUNY, USA 10:25-10:45 Robust Optimization for Decision Making under Uncertainty Florian Augustin, and Youssef M. Marzouk, Massachusetts Institute of Technology, USA
94
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
MS141
MS142
9:10 AM-10:50 AM
Reducing Communication in Sparse Linear Solvers 9:10 AM-10:50 AM
Anderson Acceleration and Applications 9:10 AM-10:50 AM
Room:251 C
Room:251 D
Room:251 E
Part of the SIAM Workshop Celebrating Diversity
As communication becomes relatively more expensive to computation on modern and future architectures, there has been much recent interest in reducing communication in linear algebra. For sparse solvers, communication-avoiding methods based on s-step iterative methods have shown great promise, but have not yet been widely adopted in practice. Hierarchical or nested solvers is another way to reduce communication and synchronization. We present recent work in these areas, with focus on preconditioned Krylov subspace solvers. Such solvers play a crucial role in CS&E.
Anderson acceleration is an algorithm for accelerating the convergence of fixed point or Picard iteration. The method was invented in 1965 to accelerate the SCF iteration in electronic structure computations and is now widely used in that field. Anderson acceleration does not require the computation or approximation of Jacobians or Jacobian-vector products, and this can be an advantage over Newton-like methods. In this minisymposium the speakers will discuss new convergence results, open questions, and applications including multiphysics coupling and radiative transport.
Organizer: Erik G. Boman
Organizer: C.T. Kelley
Sandia National Laboratories, USA
North Carolina State University, USA
9:10-9:30 Enlarged Krylov Subspace Methods for Reducing Communication Laura Grigori and Sophie Moufawad, INRIA, France; Frederic Nataf, Laboratoire Jacques-Louis Lions, France
9:10-9:30 Anderson Acceleration: Convergence Theory and Numerical Experience C.T. Kelley and Alex Toth, North Carolina State University, USA
9:35-9:55 Preconditioning Communication-Avoiding Krylov Methods Siva Rajamanickam, Sandia National Laboratories, USA
9:35-9:55 Anderson Acceleration for Parallel Applications John Loffeld and Carol S. Woodward, Lawrence Livermore National Laboratory, USA
10:00-10:20 Efficient DeflatedBased Preconditioning for the Communication-Avoiding Conjugate Gradient Method Erin C. Carson, Nicholas Knight, and James W. Demmel, University of California, Berkeley, USA
10:00-10:20 On the Performance of Anderson Acceleration for Multiphysics Problems Roger Pawlowski, Sandia National Laboratories, USA; Steven Hamilton and Mark Berrill, Oak Ridge National Laboratory, USA; Alexander R. Toth and C.T. Kelley, North Carolina State University, USA; Andrew Salinger, Sandia National Laboratories, USA
Monday, March 16
MS140 Modern Computational Modeling in Fluids
Recent advances in computational modeling in fluids have given researchers the ability to address complex hydrodynamic phenomena observed in experiments that arise in both physical fluids problems as well as more modern biological fluid applications. In this minisymposium a survey of new modeling and computational approaches, inspired by experiment, will be presented to a diverse set of challenging problems in these areas. These problems include modern computational fluid approaches to more classic problems such as surfactant spreading and vortex merger as well as important biological problems such as sperm motility and thrombus formation.
Organizer: David T. Uminsky University of San Francisco, USA
Organizer: Raegan Higgins Texas Tech University, USA
Organizer: Susan E. Minkoff University of Texas at Dallas, USA
Organizer: Stephen Wirkus Arizona State University, USA 9:10-9:30 A Computational Model of Sperm Motility Through Viscoelastic Networks Jacek Wrobel, Ricardo Cortez, and Lisa J. Fauci, Tulane University, USA 9:35-9:55 Mathematical Modeling of Blood Clot Formation Under Flow Karin Leiderman, University of California, Merced, USA 10:00-10:20 Experiment-Driven Surfactant Spreading Models Rachel Levy, Harvey Mudd College, USA 10:25-10:45 A Multi-Moment Approach to Modeling the Onset of Vortex Merger David T. Uminsky, University of San Francisco, USA
10:25-10:45 Hierarchical and Nested Krylov Methods for Extreme-Scale Computing Hong Zhang and Lois Curfman McInnes, Argonne National Laboratory, USA
10:25-10:45 Accelerating the EM Algorithm for Mixture-density Estimation Joshua H. Plasse and Homer F. Walker, Worcester Polytechnic Institute, USA
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
MS143 Parametric Model Reduction and Inverse Problems - Part III of IV 9:10 AM-10:50 AM Room:251 F For Part 2 see MS116 For Part 4 see MS169 Optimization and nonlinear inverse problems require the evaluation of a nonlinear function for a slowly changing vector of parameters. If this function involves the solution of one or more partial differential equations, the computational cost may become intractable. An important way to reduce the computational cost is the use of reduced models for the function evaluation and its derivatives. Typically, one wants to maintain the parametric dependence of the problem in the reduced model. The presentations in this minisymposium consider both applications and the theoretical background of parametric model reduction and inverse problems as well as competing stochastic approaches and their links with model reduction.
Organizer: Eric De Sturler Virginia Tech, USA
Organizer: Lars Ruthotto Emory University, USA
Organizer: Eldad Haber University of British Columbia, Canada
Organizer: Misha E. Kilmer Tufts University, USA
Organizer: Christopher A. Beattie Virginia Tech, USA
Organizer: Serkan Gugercin Virginia Tech, USA 9:10-9:30 Solution of the Full Waveform Inversion Problems via Projection based Reduced Order Models Vladimir L. Druskin, Schlumberger-Doll Research, USA; Alexander V. Mamonov, University of Texas at Austin, USA; Mikhail Zaslavsky, Schlumberger-Doll Research, USA
continued in next column
9:35-9:55 Overlapping Clustering and Ldeim in Model Reduction for Nonlinear Inversion Alexander Grimm, Serkan Gugercin, Christopher A. Beattie, and Eric De Sturler, Virginia Tech, USA; Misha E. Kilmer, Tufts University, USA 10:00-10:20 An Efficient Output Error Bound for Model Order Reduction of Parametrized Nonlinear Evolution Equations Peter Benner, Lihong Feng, and Yongjin Zhang, Max Planck Institute for Dynamics of Complex Systems, Germany 10:25-10:45 Efficiencies in Global Basis Approximation for Model Order Reduction in Diffuse Optical Tomography Meghan O’Connell and Misha E. Kilmer, Tufts University, USA; Eric De Sturler, Serkan Gugercin, and Christopher A. Beattie, Virginia Tech, USA
95
Monday, March 16
MS144 Computational Methods and Uncertainty Quantification for Inverse Problems Part I of II 9:10 AM-10:50 AM Room:254 A For Part 2 see MS170 Inverse problems require estimating parameters in a mathematical model from indirect observations. The number of unknown parameters is typically large, and obtaining a meaningful estimate is often impossible without special techniques such as regularization. Moreover, even with regularization, the resulting computational problems are often very challenging. A separate issue is that the measurements are stochastic, prior information contains uncertainty, and hence the solutions of inverse problems contain uncertainty. In some inverse problems applications, quantifying uncertainty in solutions is essential. The talks in this minisymposium will focus on computational methods both for solving inverse problems and for quantifying uncertainty in solutions. Organizer: Johnathan M. Bardsley University of Montana, USA Organizer: Aaron B. Luttman National Security Technologies, LLC, USA 9:10-9:30 An MCMC Approach to Quantifying Uncertainties in Neutron Tomography Aaron B. Luttman, Eric Machorro, and Daniel Lowe, National Security Technologies, LLC, USA 9:35-9:55 Hierarchical Bayesian Sampling for Image Reconstruction of X-Ray and Proton Radiographs Marylesa Howard, National Security Technologies, LLC, USA; Michael Fowler, MathWorks, USA; Aaron B. Luttman, National Security Technologies, LLC, USA; Margaret Hock, Columbia University, USA
continued on next page
96
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
Monday, March 16
MS144
MS145
MS146
Computational Methods and Uncertainty Quantification for Inverse Problems Part I of II 9:10 AM-10:50 AM
Task-based Scientific Computing Applications Part I of II 9:10 AM-10:50 AM
Student Days: Undergraduate Sessions Part I of II 9:10 AM-10:50 AM
Room:254 B
Room:254 C
For Part 2 see MS171 The extreme complexity of hardware platforms makes them harder and harder to program. To fully exploit such machines, the High Performance Community often uses a MPI + X (X being pthreads, OpenMP, Cuda ...) programming models. In this minisymposium, we overview an alternative solution consisting of programming at a higher level of abstractions by descrbing a scientific computing application as a sequence of tasks. Taking care of the dependences between tasks, a runtime system then processes the different tasks on the available computational units.
For Part 2 see MS172 Sponsored by the SIAG Applied Mathematics Education
continued
10:00-10:20 Methods for Accurate and Efficient Computation of the Proper-Orthogonal-Decomposition with Large Data Sets Brian Helenbrook and Fariddudin Behzad, Clarkson University, USA 10:25-10:45 Proper Orthogonal Decomposition Based Reduced Order Modeling for Real Time Monte Carlo Simulation Indika G. Udagedara and Brian Helenbrook, Clarkson University, USA; Aaron B. Luttman and Stephen Mitchell, National Security Technologies, LLC, US
Organizer: Emmanuel Agullo INRIA, France
Organizer: Hatem Ltaief King Abdullah University of Science & Technology (KAUST), Saudi Arabia 9:10-9:30 A Task-Based Sparse Direct Solver Suited for Large Scale Hierarchical/heterogeneous Architectures Pierre Ramet, LABRI, Univ Bordeaux, France 9:35-9:55 Sparse Direct Solvers on Top of a Runtime System Emmanuel Agullo, INRIA, France; Alfredo Buttari, CNRS, France; Florent Lopez, Universite Paul Sabatier, France; Abdou Guermouche, LaBRI, France 10:00-10:20 An Adaptable, Application-Aware Task-Centric Runtime System George Bosilca, University of Tennessee, Knoxville, USA 10:25-10:45 Task-Based Parallelization of the Fast Multipole Method on NVIDIA GPUs and Multicore Processors Eric F. Darve, Stanford University, USA; Emmanuel Agullo, Berenger Bramas, and Olivier Coulaud, INRIA, France; Matthias Messner, Stanford University, USA; Toru Takahashi, Nagoya University, Japan
Undergraduate Research in Applied and Computational Mathematics.
Organizer: Peter R. Turner Clarkson University, USA 9:10-9:21 Interpreting Twitter Data from World Cup Tweets Carol Sadek and Caley Johns, Wofford College, USA 9:26-9:37 Higher Dimensional Smooth Data Interpolation: Algorithmic Techniques from Computational Geometry Ariel Herbert-Voss, University of Utah, USA 9:42-9:53 Valuation of American Options and E. Coli Mutations James A. Stronz, University of Illinois at Urbana-Champaign, USA 9:58-10:09 The Effects of Chronic Wasting Disease on Pennsylvania Deer and Coyote Populations Brandon D. Thrush, Shippensburg University, USA 10:14-10:25 Application of Lévy-Flight Firefly Algorithm in Solving Several Engineering Problems Fauziah Andini Putri, Bandung Institute of Technology, Indonesia 10:30-10:41 A Bioinformatic Approach to Colorectal Cancer Research Nicolas Limogiannis and Nick Napier, Wofford College, USA
2015 SIAM Conference on Computational Science and Engineering
97
Monday, March 16
Monday, March 16
Monday, March 16
MS147
MS148
MS149
Hybrid and Multilevel Approaches to Kinetic Equations - Part I of IV 9:10 AM-10:50 AM
Reduced-order Models for PDE-constrained Optimization Problems Part I of II 9:10 AM-10:50 AM
FASTMath Solver Technologies: Advances and Applications Part I of II 9:10 AM-10:50 AM
Room:150 DE
Room:258
Organizer: Ryan G. McClarren
For Part 2 see MS174 Model reduction is an indispensable tool for simulation-based science, in problem tasks where multiple simulation requests or realtime simulation response by fast approximate models are desired. This minisymposium addresses the particular challenges of model reduction for optimization and optimal control, i.e. scenarios such as shape optimization, parameter optimization, multiobjective optimization, optimal control, feedback control or other PDE-constrained optimization problems. Such problems are relevant to many applications in science and engineering, such as fluid mechanics, solid mechanics, geophysics, electro-magnetics and acoustics.
For Part 2 see MS175 Efficient, scalable, and reliable algorithms for the solution of algebraic equations are crucial for the success of large-scale simulations. This minisymposium focuses on new developments in FASTMath solver algorithms and software, which include iterative and direct linear solution methods, nonlinear solvers, and eigensolvers, and their use in applications. We will describe efforts to increase the efficiency of algorithms and software, extensions of their usability to next generation computers, including heterogeneous architectures, mesh solver interactions as well as the impact of FASTMath solvers on large-scale multiphysics applications.
Texas A&M University, USA
Organizer: Andrea Manzoni
Organizer: Ulrike M. Yang
Organizer: Martin Frank
École Polytechnique Fédérale de Lausanne, Switzerland
Lawrence Livermore National Laboratory, USA
Organizer: Bernard Haasdonk
9:10-9:30 New Developments in hypre’s Interfaces and Solvers Ulrike Meier Yang, Lawrence Livermore National Laboratory, USA
Room:150 AB For Part 2 see MS173 In recent years, a significant amount of effort has been directed to solving kinetic equations. Generally, kinetic equations have posed a challenge due to the large phase space associated with the equations, however algorithmic advances and advances in computer hardware have made these simulations more tractable. In this minisymposium, we consider hybrid and multilevel approaches to solving kinetic equations.
Organizer: Jeffrey A. Willert Los Alamos National Laboratory, USA
Organizer: Cory Hauck Oak Ridge National Laboratory, USA
RWTH - Aachen University of Technology, Germany
Organizer: Jingmei Qiu
University of Stuttgart, Germany
University of Houston, USA 9:10-9:30 Kinetic Theory Molecular Dynamics Frank Graziani, Lawrence Livermore National Laboratory, USA
9:10-9:30 Projection-based ROMs for Parametrized PDE-constrained Optimization and Control Problems Andrea Manzoni, École Polytechnique Fédérale de Lausanne, Switzerland
9:35-9:55 Realizability in High-Order Numerical Solutions of Entropy-Based Moment Closures Graham Alldredge, RWTH Aachen University, Germany
9:35-9:55 Aeroelastic Design Optimization with Flutter Constraints and Local Rom Interpolation Youngsoo Choi, David Amsallem, and Charbel Farhat, Stanford University, USA
10:00-10:20 A High-Order / LowOrder Approach to Ocean Modeling Chris Newman, Geoff Womeldorff, Dana Knoll, and Luis Chacon, Los Alamos National Laboratory, USA
10:00-10:20 POD-G Reduced Order Models for Prediction and Control of Turbulent Flows S.S. Ravindran, University of Alabama, Huntsville, USA
10:25-10:45 Control Strategies for Multi-Agent Games Michael Herty, RWTH - Aachen University of Technology, Germany
10:25-10:45 Parameterized ReducedOrder Models for Shape Optimization of Flow Domains Jeff Borggaard, Virginia Tech, USA
9:35-9:55 Strategies for Reducing Setup Costs in Algebraic Multigrid Jonathan J. Hu and Andrey Prokopenko, Sandia National Laboratories, USA 10:00-10:20 A Distributed CPU-GPU Sparse Direct Solver Xiaoye Sherry Li, Lawrence Berkeley National Laboratory, USA; Piyush Sao and Richard Vuduc, Georgia Institute of Technology, USA 10:25-10:45 Next Generation Sparse Symmetric Factorization Mathias Jacquelin and Esmond G. Ng, Lawrence Berkeley National Laboratory, USA
98
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
MS151
MS152
High-Order Methods for Computational Fluid Dynamics - Part III of IV 9:10 AM-10:50 AM
Software Components for Integral Equation Methods Part I of IV 9:10 AM-10:50 AM
Room:259
Room:260 A
Recent Advances in a Posteriori Error Estimations and Adaptive Methods Part I of II 9:10 AM-10:50 AM
For Part 2 see MS123 For Part 2 see MS176 Novel contributions in the field of highorder numerical methods for computational fluid dynamics. Presentations will explore new high-order methods, benchmarking of existing schemes, and applications to turbulent flow problems.
For Part 2 see MS177 Due to recent advances in the underlying technology, integral equations have become highly competitive in attacking the most challenging of elliptic, globally-coupled PDE problems. However, compared to a standard off-the-shelf finite element methods, a larger amount of machinery is involved, including fast algorithms, preconditioners, and singular quadrature methods. To make this theoretically attractive family of methods practically usable, many of these components need to be readily available in software form. This minisymposium seeks to bring together practitioners and researchers with the goal of encouraging exchange of ideas and discovering opportunities for collaboration in this direction.
Monday, March 16
MS150
Organizer: Brian C. Vermeire Imperial College London, United Kingdom
Organizer: Antony Jameson Stanford University, USA
Organizer: Peter E. Vincent Imperial College London, United Kingdom 9:10-9:30 Understanding the Role of Spectral Vanishing Viscosity in High Reynolds Number Flows Rodrigo Moura, Jean-Eloi Lombard, David Moxey, Yan Bao, and Spencer Sherwin, Imperial College London, United Kingdom 9:35-9:55 A Theoretical and Computational Framework for Measure-Valued Solutions to Conservation Laws Zakerzadeh Mohammad and Georg May, RWTH Aachen University, Germany 10:00-10:20 An Implicit Les Strategy for High Order Discontinuous Galerkin Discretizations Andrea D. Beck and David Flad, University of Stuttgart, Germany; Claus-Dieter Munz, Institut fuer Aerodynamik und Gasdynamik (IAG), Germany 10:25-10:45 High-Order Finite-Volume Solution of Turbulent Aerodynamic Flows Alireza Jalali and Carl Ollivier-Gooch, University of British Columbia, Canada
Room:260 B
University College London, United Kingdom
For Part 2 see MS178 Self-adaptive numerical methods provide a powerful and automatic approach in scientific computing. In particular, Adaptive Mesh Refinement (AMR) algorithms have been widely used in computational science and engineering and have become a necessary tool in computer simulations of complex natural and engineering problems. The key ingredient for success of selfadaptive numerical methods is a posteriori error estimates that are able to accurately locate sources of global and local error in the current approximation. Talks in this minisymposium will cover some recent advances in the development and analysis of both a posteriori estimators and (convergent) adaptive schemes, as well as indicate directions of future research
Organizer: Andreas Kloeckner
Organizer: Zhiqiang Cai
Organizer: Timo Betcke
University of Illinois, USA
Purdue University, USA
9:10-9:30 BEM++ - Building Blocks for Galerkin Boundary Element Methods Timo Betcke, Simon Arridge, and Elwin van’t Wout, University College London, United Kingdom
Organizer: Shun Zhang
9:35-9:55 Resolving Uncomfortable Tradeoffs in Building Fast BoundaryElement Method Solvers: It’s Not the Hows, It’s the Whys Jaydeep Bardhan, Northeastern University, USA; Matthew G. Knepley, University of Chicago, USA 10:00-10:20 A Numerical Routine for Fast Spherical Grid Rotations Shravan Veerapaneni, University of Michigan, USA 10:25-10:45 Applications of Accelerated BEM in Aeronautics Nolwenn Balin and Benoit Lizé, Airbus, United Kingdom; Guillaume Sylvand, Airbus Operation S.A.S., France; Isabelle Terrasse, Airbus, United Kingdom
City University of Hong Kong, Hong Kong 9:10-9:30 Recovery Based a Posteriori Error Estimation for Finite Element Methods Zhiqiang Cai, Purdue University, USA 9:35-9:55 Localized H(div) RecoveryBased a Posteriori Error Estimators Xu Zhang and Zhiqiang Cai, Purdue University, USA 10:00-10:20 Robust a-Posteriori Error Estimation for Finite Element Approximation to H(curl) Problem Shuhao Cao, Pennsylvania State University, USA 10:25-10:45 Adaptive Finite Element Methods for Fourth Order Problems Jun Hu, Peking University, China
2015 SIAM Conference on Computational Science and Engineering
99
Monday, March 16
Monday, March 16
Monday, March 16
MS153
MS154
MS155
A Hierarchy of Models for Computational Neuroscience 9:10 AM-10:50 AM
Novel Spectral Approximation: Theory, Algorithms and Applications - Part I of II 9:10 AM-10:50 AM
Large Scale Computing in the Geosciences Part I of II 9:10 AM-10:50 AM
Room:151 DE
For Part 2 see MS181 Advances in scientific computing (new algorithms, open-source software, high performance computing) have made it possible to solve larger and larger problems. These advances are especially advantageous in the geosciences, where typical problems are time-dependent and three-dimensional with little or no symmetry and with important dynamics at multiple scales. Geoscientists are increasingly relying on cutting-edge software and algorithms to advance the understanding of planet Earth. This minisymposium brings together geophysicists, mathematicians, and computer scientists to discuss the latest developments at the intersection of these fields.
Room:151 AB This minisymposium will focus on a hierarchy of mathematical/physical models (from discrete ions to the continuum driftdiffusion model to a continuum spine model) for predicting neurological behavior at various levels. Numerical simulations and mathematical analyses of the models will be presented, and applied to diverse neuroscience problems like the retina, sensory processing, and neuron signaling.
Organizer: Carl L. Gardner Arizona State University, USA
Organizer: Steven M. Baer Arizona State University, USA 9:10-9:30 Continuum Spine Modeling with Application to Outer Retina Neurocircuitry Steven M. Baer, Arizona State University, USA 9:35-9:55 Simulation of the Ephaptic Effect in the Cone-Horizontal Cell Synapse of the Retina Carl L. Gardner, Arizona State University, USA 10:00-10:20 Modeling of CalciumInduced Calcium Release Dirk Gillespie, Rush University Medical Center, USA 10:25-10:45 Dendritic Coincidence Detection Enabling Wordspotting Computation Jennifer Hasler, Georgia Institute of Technology, USA
For Part 2 see MS180 Spectral and high-order methods traditionally offer high-accuracy computational results in scientific computing. Yet, recent research focuses on very different and compelling advantages for applications. These include low dispersion for wave propagation problems, near-minimal degrees of freedom for oscillatory problems and great flexibility in dealing with complicated geometries. The goal of this minisymposium is to present an overview of such contemporary research topics. Methods of recent interest include Fourier extensions or continuations, radial basis functions and redundant frame discretizations. Challenges involve obtaining fast transforms and wellconditioned discretizations, or overcoming ill-conditioning in methods involving redundancy. The focus ranges from theory to applications.
Room:151 G
Organizer: Sander Rhebergen University of Oxford, United Kingdom
Organizer: Andrew J. Wathen University of Oxford, United Kingdom
Organizer: Richard F. Katz
Organizer: Daan Huybrechs
University of Oxford, United Kingdom
Katholieke Universiteit Leuven, Belgium
9:10-9:30 Three-Field BlockPreconditioners for Models of Coupled Magma/mantle Dynamics Sander Rhebergen, University of Oxford, United Kingdom; Garth Wells, University of Cambridge, United Kingdom; Andrew J. Wathen and Richard F. Katz, University of Oxford, United Kingdom; Laura Alisic and John Rudge, University of Cambridge, United Kingdom
9:10-9:30 Fast Spectral PDE Solvers for Complex Structures: the FourierContinuation Method Oscar P. Bruno, California Institute of Technology, USA 9:35-9:55 Frame Theoretic Convolutional Gridding Anne Gelb, Arizona State University, USA; Guohui Song, Clarkson University, USA 10:00-10:20 Optimized Fourier Continuation Methods Mark Lyon, University of New Hampshire, USA
9:35-9:55 Parallel and Adaptive Mantle Convection Simulation in Aspect Timo Heister, Clemson University, USA; Wolfgang Bangerth, Texas A&M University, USA
10:25-10:45 A New Radial Basis Functions (RBF)-based Frame Method to Bypass the Runge Phenomenon Cecile M. Piret, Université Catholique de Louvain, Belgium
10:00-10:20 HPC Finite Elements for Nonlinear Stokes Flow Dave A. May, ETH Zürich, Switzerland; Jed Brown, Argonne National Laboratory, USA and University of Colorado Boulder, USA 10:25-10:45 Large-Scale Forward and Inverse Numerical Simulations of Crustal and Lithospheric-Scale Deformation Boris Kaus, Anton Popov, and Tobias Baumann, Johannes Gutenberg-Universität, Mainz, Germany
100
Monday, March 16 Coffee Break 10:50 AM-11:20 AM Room:255
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
PD3 Panel The Future of CSE as a Discipline 12:15 PM-1:30 PM Room:355 D
IP6 Scaling Open Systems for Future Computational Challenges 11:20 AM-12:05 PM Room:355 Chair: Anders Logg, Chalmers University of Technology, Sweden Computational models are changing rapidly, partially in response to growing data size and advances in high-performance computing. Open approaches are well suited to this dynamic environment as they provide agile responses to complex, evolving code, and support the greater goal of ensuring reproducible science. This presentation introduces some open initiatives addressing big data and HPC and the role that software architectures and processes plays in advancing scientific computation. Also discussed are emerging trends including competitive challenges and active publications that will likely play an important role in the creation, development and deployment of computational software.
Will Schroeder Kitware, Inc., USA
Chair: David E. Keyes, King Abdullah University of Science & Technology (KAUST), Saudi Arabia Over the past two decades Computational Science and Engineering (CSE) has penetrated the academy, with prominent roles in advancing research and providing interdisciplinary education. However, a combination of disruptive developments -- including extreme-scale computing, big data, and a significant diversification of the applications of CSE in science, technology, and society -- is redefining the scope and reach of CSE. This forward-looking panel will explore the future of CSE in a broad academic context. What new “grand challenges” may drive progress in CSE? How can CSE shape the future of new application fields such as computational medicine and biology, computational geoscience, and materials science? Are there opportunities to extend CSE to new areas such as social network analysis, cybersecurity and the social sciences, with mathematics-based large-scale computing rapidly becoming of crucial importance in almost all areas of society? Is the CSE paradigm and focus sufficiently unique and significant to warrant separate programs, graduate degrees, academic departments, and funding streams?
Panelists: Lunch Break 12:05 PM-1:30 PM Attendees on their own
Lorena Barba Boston University, USA Wolfgang Marquardt RWTH Aachen University, Germany J. Tinsley Oden University of Texas at Austin, USA Padma Raghavan Pennsylvania State University, USA Ed Seidel University of Illinois at UrbanaChampaign, USA
Monday, March 16
MS156 Featured Minisymposium: Physics-compatible Numerical Methods 1:30 PM-3:10 PM Room:355 Physics-compatible numerical methods are methods that aim to preserve key mathematical and physical properties of continuum physics models in their finitedimensional algebraic representations. They include methods which preserve prop- erties such as energy, monotonicity, maximum principles, symmetries, and involutions of the continuum models. Examples are mimetic methods for spatial discretizations, variational and geometric integrators, conservative finite-volume and finite- element methods, etc. Research on physics-compatible numerical methods is rapidly becoming a major research thrust across multiple disciplines within the broader area of computational science and engineering. Our principal goal in arranging this minisymosium is to give samples of this flourishing field.
Organizer: Mikhail Shashkov Los Alamos National Laboratory, USA 1:30-1:50 Mimetic Finite-Difference Methods Mikhail Shashkov, Los Alamos National Laboratory, USA 1:55-2:15 A High-Order/Low-Order Exponentially-Convergent IMC Method Simon Bolding and Jim E. Morel, Texas A&M University, USA; Robert B. Lowrie, Los Alamos National Laboratory, USA 2:20-2:40 Multiphysics Lagrangian/ Eulerian Modeling and deRham Complex Based Algorithms Allen C. Robinson, Sandia National Laboratories, USA 2:45-3:05 A Sub Cell Dynamics Based Closure Model for Multimaterial Arbitrary Lagrangian Eulerian Codes Andrew J. Barlow, Atomic Weapons Establishment, United Kingdom
2015 SIAM Conference on Computational Science and Engineering
101
Monday, March 16
Monday, March 16
Monday, March 16
MS157
MS158
MS159
Cut Cells: Algorithms and Applications - Part II of II 1:30 PM-3:10 PM
Recent Advances in Bayesian Methods for Computational Science and Engineering - Part II of II 1:30 PM-3:10 PM
High-dimensional Approximation and Integration: Analysis and Computation - Part IV of V 1:30 PM-3:10 PM
Room:250 A
Room:250 B
For Part 1 see MS132 This minisymposium highlights the most recent impact of Bayesian inference for uncertainty quantification, statistical modeling, experimental design, and optimization. We discuss novel approximation techniques bringing with them a myriad of new potential applications from medicine to engineering, finance, and biology. The minisymposium brings together researchers from diverse application domains to introduce and transfer recent contributions, as well as to discuss open challenges due to increasing demand for efficient techniques.
Room:355 A For Part 1 see MS130 Cut cell methods solve PDEs in complex geometry using a regular Cartesian grid with special treatment for cells that intersect the boundary. These methods are attractive because they easily represent complex and moving geometries. Cut cell methods have been used with finite difference, volume and element discretizations, across a wide range of problems, including aerodynamics, meteorology, oceanography, and others. Challenges include gaps in theoretical underpinnings; issues due to arbitrarily small cells; and the programming complexity that arises with cell merging and other stabilization techniques. This minisymposium will highlight recent algorithmic advances, and include a variety of science domain applications. Lawrence Berkeley National Laboratory, USA
University of California, Santa Barbara, USA
Organizer: Marsha Berger
Organizer: James Beck
For Part 3 see MS133 For Part 5 see MS185 This minisymposium focuses on the fundamental problem of how to accurately approximate solutions of both forward and inverse high-dimensional functions. Predicting the behavior of complex phenomena relies on constructing solutions in terms of high dimensional spaces, particularly in the case when the input data (coefficients, forcing terms, initial and boundary conditions, geometry) are affected by large amounts of uncertainty. The resulting explosion in computational effort is a symptom of the curse of dimensionality and this symposium aims at exploring breakthroughs in sparse sampling and representations, nonlinear and greedy approximations, compressed sensing, multilevel methods, and ”best-N-term” approximations.
California Institute of Technology, USA
Organizer: Guannan Zhang
1:30-1:50 On the Use of Particle Based Methods for the Real Time Identification and Control of Nonlinear Dynamical Systems Eleni Chatzi, ETH Zürich, Switzerland
Oak Ridge National Laboratory, USA
1:55-2:15 Bayesian Uncertainty Quantification and Propagation for Molecular Dynamic Simulations in Nanoscale Fluid Mechanics Petros Koumoutsakos and Panagiotis Angelikopoulos, ETH Zürich, Switzerland; Costas Papadimitriou, University of Thessaly, Greece
Florida State University, USA
Organizer: Hans Johansen
Courant Institute of Mathematical Sciences, New York University, USA 1:30-1:50 A Higher-Order Cut Cell Finite Volume Method For AdvectionDiffusion Dharshi Devendran and Hans Johansen, Lawrence Berkeley National Laboratory, USA 1:55-2:15 A Mixed Explicit Implicit Time Stepping Scheme for Cartesian Embedded Boundary Meshes Sandra May, ETH Zürich, Switzerland; Marsha Berger, Courant Institute of Mathematical Sciences, New York University, USA 2:20-2:40 Inverse Lax-Wendroff Procedure for Numerical Boundary Conditions of Hyperbolic Equations Chi-Wang Shu, Brown University, USA 2:45-3:05 Representing Topography in Earth System Models with Porous Barriers Alistair Adcroft, Princeton University, USA
Organizer: AlbertoGiovanni Busetto
2:20-2:40 Sequentially Constrained Monte Carlo Dave A. Campbell, Simon Fraser University, Canada; Shirin Golchi, Columbia University, USA 2:45-3:05 Computationally Efficient Tools for Bayesian Uncertainty Quantification and Propagation in Structural Dynamics Costas Papadimitriou, University of Thessaly, Greece; Panagiotis Angelikopoulos, Panagiotis Hadjidoukas, and Petros Koumoutsakos, ETH Zürich, Switzerland
Organizer: Albert Cohen Université Pierre et Marie Curie, France
Organizer: Max Gunzburger Organizer: Clayton G. Webster Oak Ridge National Laboratory, USA 1:30-1:50 Option Pricing and the Anova Decomposition of a Function of An Infinite Number of Variables Ian H. Sloan and Frances Y. Kuo, University of New South Wales, Australia; Michael Griebel, University of Bonn, Germany 1:55-2:15 Application of Quasi-Monte Carlo Methods to PDEs with Random Coefficients Frances Y. Kuo, University of New South Wales, Australia 2:20-2:40 Adaptive Monte Carlo and Quasi-Monte Carlo Integration Fred J. Hickernell, Lan Jiang, and Antoni Luís Jiménez Rugama, Illinois Institute of Technology, USA continued on next page
102
Monday, March 16
MS159
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
MS160
High-dimensional Approximation and Integration: Analysis and Computation - Part IV of V 1:30 PM-3:10 PM
Resilience in Numerical Simulations and Algorithms at Extreme Scale: Part IV of IV - Applications 1:30 PM-3:10 PM
continued
Room:250 C
2:45-3:05 H-Matrix Accelerated Second Moment Analysis for Potentials with Rough Correlation Helmut Harbrecht, Juergen Doelz, and Michael Peters, Universität Basel, Switzerland; Christoph Schwab, ETH Zürich, Switzerland
For Part 3 see MS134 The advent of extreme scale machines will require the use of parallel resources at an unprecedented scale, probably leading to a high rate of hardware faults. Handling fully these faults at the computer system level may have a prohibitive cost. High performance computing applications that aim at exploiting all these resources will thus need to be resilient, in this minisymposium different complementary approaches and methods will be presented to possibly address this key aspect of extreme scale computing.
Organizer: Keita Teranishi Sandia National Laboratories, USA
Organizer: Luc Giraud INRIA, France
Organizer: Emmanuel Agullo INRIA, France
Organizer: Michael Heroux Sandia National Laboratories, USA 1:30-1:50 Adaptive Determination of Optimal Multilevel Monte Carlo parameters in the Presence of Failures Peter Arbenz and Stefan Pauli, ETH Zürich, Switzerland 1:55-2:15 Spatial Decomposition for Resilient Extreme-Scale Scientific Simulations Francesco Rizzi, Khachik Sargsyan, Karla Morris, and Cosmin Safta, Sandia National Laboratories, USA; Paul Mycek and Omar M. Knio, Duke University, USA; Olivier LeMaitre, LIMSI-CNRS, France; Habib N. Najm and Bert J. Debusschere, Sandia National Laboratories, USA
continued in next column
2:20-2:40 Hierarchical Resilience for Structured AMR Anshu Dubey, Lawrence Berkeley National Laboratory, USA; Hajime Fujita and Zachary Rubenstein, University of Chicago, USA; Brian Van Straalen, Lawrence Berkeley National Laboratory, USA; Andrew A. Chien, University of Chicago and Argonne National Laboratory, USA 2:45-3:05 Resilience Properties of Gossip-Style Algorithms Wilfried N. Gansterer, Gerhard Niederbrucker, Michael Moldaschl, and Karl Prikopa, University of Vienna, Austria
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
MS161
MS162
Model Error Assessment in Computational Physical Models - Part II of II 1:30 PM-3:10 PM
Advances in Time-stepping Methods - Part II of V 1:30 PM-3:10 PM
Room:250 D
For Part 1 see MS136 For Part 3 see MS188 Time-stepping methods are numerical methods for the time evolution of ODEs, DAEs, and PDEs in the method-of-lines framework. Many applications require specialized time-stepping methods in order to enhance efficiency or capture particular theoretical properties such as positivity, symplecticity, and strong-stabilitypreservation or practical properties such as parallelization on emerging high-performance architectures, component (e.g., multirate or local time-stepping) and physics (IMEX or semi-implicit) splittings based on sub-system properties, and reliable error estimation and control. The talks in this minisymposium will describe recent novel developments in time-stepping methods and demonstrate their benefits in practical applications.
For Part 1 see MS135 While parametric uncertainty quantification is performed routinely in computational science and engineering, model error assessment is still largely lacking and faces significant challenges that are yet to be resolved. For example, it is unclear how to properly disambiguate data noise from model errors. Recently there has been growing interest and some development towards the quantification of model errors, i.e. the inherent discrepancies from the truth that cannot be reduced within the model assumptions. This minisymposium focuses on state-of-the-art and novel methods for quantification of model errors. It highlights challenges and introduces a range of available methods to the computational science community, hoping to contribute significantly to improving predictive fidelity of computational physical models across a range of disciplines.
Organizer: Khachik Sargsyan Sandia National Laboratories, USA
Organizer: Habib N. Najm Sandia National Laboratories, USA 1:30-1:50 Uncertainty in Reynolds Stress Closures for Turbulent Flow Calculations Gianluca Iaccarino and Michael A. Emory, Stanford University, USA; Catherine Gorle, University of Antwerp, Belgium 1:55-2:15 Eddy Viscosity Model Selection for Transonic Turbulent Flows Using Shrinkage Regression Lawrence Dechant, Sophia Lefantzi, Jaideep Ray, and Srinivasan Arunajatesan, Sandia National Laboratories, USA 2:20-2:40 Estimating a Model Discrepancy Term for the Community Land Model Using Latent Heat and Runoff Observations Jaideep Ray and Laura Swiler, Sandia National Laboratories, USA; Maoyi Huang and Zhangshuan Hou, Pacific Northwest National Laboratory, USA 2:45-3:05 Formulation and Calibration of a Stochastic Model Form Error Representation for Rans Robert D. Moser, Todd Oliver, and Bryan Reuter, University of Texas at Austin, USA
Room:250 E
Organizer: Emil M. Constantinescu Argonne National Laboratory, USA
Organizer: Michael Guenther Bergische Universität, Germany
Organizer: Adrian Sandu Virginia Tech, USA
Organizer: Raymond J. Spiteri University of Saskatchewan, Canada 1:30-1:50 Optimal Convergence Rates of Co-Simulation Using Fine Structure Analysis Andreas Bartel and Kai Gausling, University of Wuppertal, Germany; Sebastian Schöps, TU Darmstadt, Germany 1:55-2:15 Trigonometric Integration Methods in Circuit Simulation Hans-Georg Brachtendorf and Kai Bittner, University of Applied Science Hagenberg, Austria
continued in next column
103
2:20-2:40 Fast Time-Domain Simulation for Reliable Fault Detection E. Jan W. ter Maten, Bergische Universität Wuppertal, Germany; Bratislav Tasic, Jos J. Dohmen, Theo G.J. Beelen, and Rick Janssen, NXP Semiconductors, Research, The Netherlands; Wil Schilders, Eindhoven University of Technology, Netherlands; Michael Guenther, Bergische Universität, Germany 2:45-3:05 Modelling Transmission Power Systems with the Implicit DAE Solver, IDA Philip Top, Carol S. Woodward, and Alan Hindmarsh, Lawrence Livermore National Laboratory, USA
104
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
MS164
MS165
Sparse Matrix Factorization on GPU's 1:30 PM-3:10 PM
UQ in Large Scale Computing - Part II of II 1:30 PM-3:10 PM
Room:250 F
Room:251 A
Many problems of engineering interest (structural analysis, circuit simulation, statistics, etc.) require the factorization and solution of large sparse linear systems. While today’s GPUs can provide several teraflops of performance in a desktop configuration, the use of GPUs for direct sparse factorization has been limited due to the very irregular nature of the problem. Recently researchers have been making significant progress in accelerating direct sparse factorization methods such as QR, LU and Cholesky, and LDLT on GPUs. The purpose of this minisymposium is to bring together researchers studying algorithms for accelerating direct sparse factorization on GPUs.
For Part 1 see MS138 The move towards ever-larger scale computing is bringing both challenges and opportunities to the field of uncertainty quantification. On the one hand, the increase in computing power helps to offset the cost of UQ, even for complex applications. On the other hand, communication bottlenecks, increased frequencies of soft and hard faults, and limits on memory create computational challenges for UQ algorithms. Furthermore, the relative sparsity of validation data for high-dimensional, complex applications puts strenuous demands on UQ to assess the predictive fidelity of the simulations. This minisymposium brings together talks that address these challenges.
Inverse Analysis and Uncertainty Quantification in Fluid Mechanics Part II of II 1:30 PM-3:10 PM
Organizer: Steven C. Rennich
Organizer: Bert J. Debusschere
NVIDIA, USA
Sandia National Laboratories, USA
Organizer: Timothy A. Davis
Organizer: Eric Phipps
Texas A&M University, USA
Sandia National Laboratories, USA
1:30-1:50 Accelerating the Supernodal Sparse Cholesky Factorization on GPUs Steven C. Rennich, NVIDIA, USA; Timothy A. Davis, Texas A&M University, USA; Darko Stosic, NVIDIA, USA
1:30-1:50 Information-Theoretic Tools for Uncertainty Quantification of High Dimensional Stochastic Models. Paul Dupuis, Brown University, USA; Markos Katsoulakis and Yannis Pantazis, University of Massachusetts, Amherst, USA; Petr Plechac, University of Delaware, USA
Organizer: Paul Bauman
1:55-2:15 Highly Scalable Hierarchical Sampling Algorithms for Gaussian Random Fields Panayot Vassilevski and Umberto E. Villa, Lawrence Livermore National Laboratory, USA
State University of New York, Buffalo, USA
Monday, March 16
MS163
1:55-2:15 Sparse Communication Avoiding Pivoting and GPUs Jonathan Hogg and Jennifer Scott, Rutherford Appleton Laboratory, United Kingdom 2:20-2:40 Sparse QR Factorization on Heterogenous Platforms with Multiple GPUs Mohamed Gadou, University of Florida, USA; Timothy A. Davis, Texas A&M University, USA; Sanjay Ranka, University of Florida, USA 2:45-3:05 GLU: LU Re-Factorization on the GPU Maxim Naumov, Sharan Chetlur, and Lung Sheng Chien, NVIDIA, USA
2:20-2:40 Use of Parallel MCMC Methods with the Community Land Model Jaideep Ray and Laura Swiler, Sandia National Laboratories, USA; Maoyi Huang and Jason Hou, Pacific Northwest National Laboratory, USA
Room:251 B For Part 1 see MS139 Advances in computing resources and algorithms have enabled simulations of engineered systems of unprecedented complexity. Increasingly, these simulations are being used to inform important design and operational decisions. In this context, providing defensible uncertainty estimates for computed quantities of interest is critical. The aim of this minisymposium is to bring together researchers developing approaches for quantifying uncertainties, including those due to uncertain parameters, inadequate physical models, and uncertain or sparse experimental data, in fluid mechanics applications. There is broad interest in both statistical approaches and fluids mechanics applications for participation in this minisymposium. State University of New York, Buffalo, USA
Organizer: Todd Oliver University of Texas at Austin, USA
Organizer: David Salac University of Buffalo, SUNY, USA
Organizer: Abani K. Patra 1:30-1:50 Calibration of the SpalartAllmaras Turbulence Model for Blunt Body Re-Entry Vehicle Flows Using DNS Data Robert D. Moser, Todd Oliver, Victor Topalian, and Rhys Ulerich, University of Texas at Austin, USA 1:55-2:15 Predictive Uncertainty Quantification of An Ablating Entry Vehicle Heatshield Roy Stogner, University of Texas at Austin, USA
2:45-3:05 A Fast N-body Algorithm for Kernel Sums in High Dimensions George Biros and Bill March, University of Texas at Austin, USA; Bo Xiao, Georgia Institute of Technology, USA continued on next page
2015 SIAM Conference on Computational Science and Engineering 2:20-2:40 Construction of Gaussian Surrogate Process Using Numerical and Modeling Error Uncertainty Hossein Aghakhani and Abani K. Patra, State University of New York, Buffalo, USA; Elaine Spiller, Marquette University, USA 2:45-3:05 Quantifying the Impact of Numerical Errors Along with Other Uncertainties on Probabilistic Hazard Mapping Elaine Spiller, Marquette University, USA; Hossein Aghakhani, State University of New York, Buffalo, USA
Monday, March 16
MS166 Computational Science for Current Multidisciplinary Research Problems 1:30 PM-3:10 PM Room:251 C Part of the SIAM Workshop Celebrating Diversity Everyday scientists conduct research in an effort to contribute to the vast array of scientific disciplines. More and more, computational science is being used as part of these efforts. This minisymposium will highlight the outstanding research being conducted using computational science, with applications encompassing several disciplines. Specifically, as part of the Workshop Celebrating Diversity, the session will highlight the work of female scientists.
Organizer: Talea Mayo Princeton University, USA
Organizer: Talea Mayo Princeton University, USA
Organizer: Susan E. Minkoff University of Texas at Dallas, USA
Organizer: Stephen Wirkus Arizona State University, USA
Organizer: Raegan Higgins Texas Tech University, USA 1:30-1:50 Optimization of Computational Simulation Set for Quantification of Hurricane Surge Extreme-Value Statistics Jennifer L. Irish, Virginia Tech, USA 1:55-2:15 Mathematical Modeling of Gliomas: Implications for Interpreting Therapeutic Efficacy Through Imaging Andrea Hawkins-Daarud and Russell Rockne, Northwestern University, USA; David Corwin, LaunchPad Lab, USA; Alexander R.A. Anderson, H. Lee Moffitt Cancer Center & Research Institute, USA; Paul Kinahan, University of Washington, USA; Kristin R. Swanson, Northwestern University, USA 2:20-2:40 Random Matrix Models for the Representation of Model Inadequacy: A Case Study in Chemical Kinetics Rebecca Morrison and Robert D. Moser, University of Texas at Austin, USA 2:45-3:05 Hurricane Storm Surge Risk Analysis for the US North Atlantic Coast Talea Mayo, and Ning Lin, Princeton University, USA
105
Monday, March 16
MS167 See Tuesday 4:25 PM
106
Monday, March 16
MS168 Inverse Problems and Data Assimilation - Part I of VI 1:30 PM-3:10 PM Room:251 E For Part 2 see MS194 The aim of this minisymposium is to document recent mathematical developments in the field of inverse problems and data assimilation based on theoretical and numerical grounds that are relevant for various scientific and real life applications. Topics for the minisymposium include but are not limited to: 1) deterministic and statistical approaches to inversion problems and data assimilation; 2) reduced order modeling inversion framework; 3) treatment of nonlinear and non-smooth processes; 4) sensitivity analysis and its application to adaptive (targeting) observation; 5) operational data assimilation systems; 6) uncertainties impact studies; 7) automatic tools to support inversion and data assimilation methodologies.
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
MS169 Parametric Model Reduction and Inverse Problems Part IV of IV 1:30 PM-3:10 PM Room:251 F For Part 3 see MS143 Optimization and nonlinear inverse problems require the evaluation of a nonlinear function for a slowly changing vector of parameters. If this function involves the solution of one or more partial differential equations, the computational cost may become intractable. An important way to reduce the computational cost is the use of reduced models for the function evaluation and its derivatives. Typically, one wants to maintain the parametric dependence of the problem in the reduced model. The presentations in this minisymposium consider both applications and the theoretical background of parametric model reduction and inverse problems as well as competing stochastic approaches and their links with model reduction.
Organizer: Eric De Sturler
Organizer: Razvan Stefanescu
Virginia Tech, USA
Virginia Tech, USA
Organizer: Lars Ruthotto
Organizer: Adrian Sandu
Emory University, USA
Virginia Tech, USA
Organizer: Eldad Haber
Organizer: Ionel M. Navon
University of British Columbia, Canada
Florida State University, USA
Organizer: Misha E. Kilmer
Organizer: Dacian N. Daescu
Tufts University, USA
Portland State University, USA
Organizer: Christopher A. Beattie
1:30-1:50 Nonlinear Model Order Reduction Using Pod/DEIM 4-D Var with Trust Region Applied to a Spherical Shallow Water Equations Model Ionel M. Navon, Florida State University, USA; Fangxin Fang, Imperial College London, United Kingdom; Juan Du, Academia Sinica, China
Virginia Tech, USA
1:55-2:15 Aposteriori Error Estimates and Adaptive Reduced Order Modeling Data Assimilation Razvan Stefanescu and Adrian Sandu, Virginia Tech, USA 2:20-2:40 Interpolatory Model Reduction for Nonlinear Inversion Serkan Gugercin, Virginia Tech, USA 2:45-3:05 Reduced Order Modelling for Fluid-Structure Interaction Problems Francesco Ballarin, Politecnico di Milano, Italy; Gianluigi Rozza, SISSA, International School for Advanced Studies, Trieste, Italy
Organizer: Serkan Gugercin Virginia Tech, USA 1:30-1:50 Numerical Stability Issues in H_2 approximation Methods Zlatko Drmac, University of Zagreb, Croatia; Christopher A. Beattie and Serkan Gugercin, Virginia Tech, USA 1:55-2:15 Stochastic Approach to Nonlinear Inversion Combining Simultaneous Random and Deterministic Sources Selin Sariaydin, Eric De Sturler, and Serkan Gugercin, Virginia Tech, USA; Misha E. Kilmer, Tufts University, USA
continued in next column
2:20-2:40 Certified Reduced Basis Model Reduction for Maxwell’s Equations Martin W. Hess and Peter Benner, Max Planck Institute for Dynamics of Complex Systems, Germany 2:45-3:05 Parameter Estimation for Inverse Problems Justin Krueger, Virginia Tech, USA
2015 SIAM Conference on Computational Science and Engineering
107
Monday, March 16
Monday, March 16
Monday, March 16
MS170
MS171
MS172
Computational Methods and Uncertainty Quantification for Inverse Problems - Part II of II 1:30 PM-3:10 PM
Task-based Scientific Computing Applications Part II of II 1:30 PM-3:10 PM
Student Days: Undergraduate Sessions Part II of II 1:30 PM-3:10 PM
Room:254 B
Room:254 C
Room:254 A
For Part 1 see MS145 The extreme complexity of hardware platforms makes them harder and harder to program. To fully exploit such machines, the High Performance Community often uses a MPI + X (X being pthreads, OpenMP, Cuda ...) programming models. In this minisymposium, we overview an alternative solution consisting of programming at a higher level of abstractions by descrbing a scientific computing application as a sequence of tasks. Taking care of the dependences between tasks, a runtime system then processes the different tasks on the available computational units.
For Part 1 see MS146 Sponsored by the SIAG Applied Mathematics Education
For Part 1 see MS144 Inverse problems require estimating parameters in a mathematical model from indirect observations. The number of unknown parameters is typically large, and obtaining a meaningful estimate is often impossible without special techniques such as regularization. Moreover, even with regularization, the resulting computational problems are often very challenging. A separate issue is that the measurements are stochastic, prior information contains uncertainty, and hence the solutions of inverse problems contain uncertainty. In some inverse problems applications, quantifying uncertainty in solutions is essential. The talks in this minisymposium will focus on computational methods both for solving inverse problems and for quantifying uncertainty in solutions.
Organizer: Johnathan M. Bardsley University of Montana, USA
Organizer: Aaron B. Luttman National Security Technologies, LLC, USA 1:30-1:50 Using Numerical Optimization Methods for Sampling in Inverse Problems Johnathan M. Bardsley, University of Montana, USA 1:55-2:15 Point Spread Reconstruction from the Image of a Sharp Edge: Computation and Uncertainty Quantification Kevin Joyce and Johnathan M. Bardsley, University of Montana, USA; Aaron B. Luttman, National Security Technologies, LLC, USA; Peter Golubstov, Moscow State University, Russia 2:20-2:40 Constrained Iterative Solver for Sparse Unmixing and Deblurring of Hyperspectral Images Sebastian Berisha, University of Pennsylvania, USA; James G. Nagy, Emory University, USA; Robert Plemmons, Wake Forest University, USA 2:45-3:05 Statistical Tests for Total Variation Regularization Parameter Selection Jodi Mead, Boise State University, USA
Organizer: Emmanuel Agullo INRIA, France
Organizer: Hatem Ltaief King Abdullah University of Science & Technology (KAUST), Saudi Arabia 1:30-1:50 Task Based Programming with Pycompss: Leveraging Python in Parallel Platforms Rosa M. Badia, Barcelona Supercomputing Center, Spain 1:55-2:15 A Task-Based Computational Astronomy Application Hatem Ltaief, King Abdullah University of Science & Technology (KAUST), Saudi Arabia 2:20-2:40 Coarse Grained Task-Based Parareal Parallel-In-Time Applications in Fusion Energy Wael R. Elwasif, Oak Ridge National Laboratory, USA; Debasmita Samaddar, UK Atomic Energy Authority, United Kingdom 2:45-3:05 Applications at Airbus Group of a Task-Based H-Matrix Solver Guillaume Sylvand, Airbus Operation S.A.S., France; Benoit Lize, Airbus, United Kingdom
Undergraduate Research in Applied and Computational Mathematics.
Organizer: Peter R. Turner Clarkson University, USA 1:30-1:41 Bounds on Electrical Fields in Two-Component Inhomogeneous Bodies Zoe Koch, Michael Primrose, and Michael Zhao, University of Utah, USA 1:46-1:57 Pymethyl: A Bioinformatic Approach to Methylation Patterns and their Epigenetic Effects on Risk of Breast Cancer Cody Watson, Wofford College, USA 2:02-2:13 Estimation of Unmodeled Gravitational Wave Transients: an Application of Spline Based Regression and Particle Swarm Optimization Calvin Leung, Harvey Mudd College, USA 2:18-2:29 Modeling Bull Sperm Motility Using Image Processing Linan Zhang, Worcester Polytechnic Institute, USA 2:34-2:45 Mean Squared Displacement and Mean First Passage Time in Fluids with Memory Michael Senter, University of Utah, USA 2:50-3:01 Persistent Random Walk of Microorganisms in a Porous Medium Grace Lim, California State Polytechnic University, Pomona, USA; Aden Forrow, Massachusetts Institute of Technology, USA
108
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
MS174
MS175
Reduced-order Models for PDE-constrained Optimization Problems Part II of II 1:30 PM-3:10 PM
FASTMath Solver Technologies: Advances and Applications Part II of II 1:30 PM-3:10 PM
Room:150 DE
Room:258
Oak Ridge National Laboratory, USA
For Part 1 see MS148 Model reduction is an indispensable tool for simulation-based science, in problem tasks where multiple simulation requests or realtime simulation response by fast approximate models are desired. This minisymposium addresses the particular challenges of model reduction for optimization and optimal control, i.e. scenarios such as shape optimization, parameter optimization, multiobjective optimization, optimal control, feedback control or other PDE- constrained optimization problems. Such problems are relevant to many applications in science and engineering, such as fluid mechanics, solid mechanics, geophysics, electro-magnetics and acoustics.
For Part 1 see MS149 Efficient, scalable, and reliable algorithms for the solution of algebraic equations are crucial for the success of large-scale simulations. This minisymposium focuses on new developments in FASTMath solver algorithms and software, which include iterative and direct linear solution methods, nonlinear solvers, and eigensolvers, and their use in applications. We will describe efforts to increase the efficiency of algorithms and software, extensions of their usability to next generation computers, including heterogeneous architectures, mesh solver interactions as well as the impact of FASTMath solvers on large-scale multiphysics applications.
Organizer: Ryan G. McClarren
Organizer: Andrea Manzoni
Organizer: Ulrike M. Yang
École Polytechnique Fédérale de Lausanne, Switzerland
Lawrence Livermore National Laboratory, USA
RWTH - Aachen University of Technology, Germany
Organizer: Bernard Haasdonk
Organizer: Jingmei Qiu
1:30-1:50 Reduced Order Models for Nonlinear PDE-Constrained Optimization Problems in Fluid Dynamics Federico Negri, Andrea Manzoni, and Alfio Quarteroni, École Polytechnique Fédérale de Lausanne, Switzerland
1:30-1:50 FASTMath Unstructured Mesh (MOAB) Solver (PETSc) Interactions Vijay Mahadevan, Iulian Grindeanu, and Barry F. Smith, Argonne National Laboratory, USA
Monday, March 16
MS173 Hybrid and Multilevel Approaches to Kinetic Equations - Part II of IV 1:30 PM-3:10 PM Room:150 AB For Part 1 see MS147 For Part 3 see MS199 In recent years, a significant amount of effort has been directed to solving kinetic equations. Generally, kinetic equations have posed a challenge due to the large phase space associated with the equations, however algorithmic advances and advances in computer hardware have made these simulations more tractable. In this minisymposium, we consider hybrid and multilevel approaches to solving kinetic equations.
Organizer: Jeffrey A. Willert Los Alamos National Laboratory, USA
Organizer: Cory Hauck
Texas A&M University, USA
Organizer: Martin Frank
University of Houston, USA 1:30-1:50 A Deterministic-Particle Transport Solver for Scale-Bridging Simulation of Thermal Radiative Transfer Hyeongkae Park, Los Alamos National Laboratory, USA 1:55-2:15 Practical Numerical Methods for Solving the Boltzmann Transport Equation in Nuclear Reactor Analysis Rodolfo Ferrer, Studsvik Scandpower, Inc., USA 2:20-2:40 A Hierarchy of Hybrid Numerical Methods for Multi-Scale Kinetic Equations Thomas Rey, University of Lille, France; Francis Filbet, University of Lyon 1, France 2:45-3:05 On a New Class of SemiLagrangian Schemes for Kinetic Equations Giacomo Dimarco, Università degli Studi di Ferrara, Italy
University of Stuttgart, Germany
1:55-2:15 Application of Discrete Empirical Interpolation Method to Reduced Order Modeling of Nonlinear Parametric Systems Harbir Antil, George Mason University, USA 2:20-2:40 Reduced Basis Method for Hamilton-Jacobi-Bellman Equations Sebastian Steck and Karsten Urban, University of Ulm, Germany 2:45-3:05 HJB-POD Feedback Control for Advection-Diffusion Equations Alessandro Alla and Michael Hinze, University of Hamburg, Germany
1:55-2:15 Application of Algebraic Multigrid (PETSc) for Block Structured Adaptive Mesh Refinement Applications (Chombo) Mark Adams, Lawrence Berkeley National Laboratory, USA 2:20-2:40 Scalable Adaptive ImEx Integration with ARKode and HYPRE Daniel R. Reynolds, Southern Methodist University, USA 2:45-3:05 Solvers and Error Control for Atmospheric Column Physics Jed Brown, Argonne National Laboratory, USA and University of Colorado Boulder, USA
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
MS176 High-Order Methods for Computational Fluid Dynamics - Part IV of IV 1:30 PM-3:10 PM
Monday, March 16
MS177 Software Components for Integral Equation Methods Part II of IV 1:30 PM-3:10 PM
Room:259
Room:260 A
For Part 3 see MS150 Novel contributions in the field of highorder numerical methods for computational fluid dynamics. Presentations will explore new high-order methods, benchmarking of existing schemes, and applications to turbulent flow problems.
For Part 1 see MS151 For Part 3 see MS203 Due to recent advances in the underlying technology, integral equations have become highly competitive in attacking the most challenging of elliptic, globally-coupled PDE problems. However, compared to a standard off-the-shelf finite element methods, a larger amount of machinery is involved, including fast algorithms, preconditioners, and singular quadrature methods. To make this theoretically attractive family of methods practically usable, many of these components need to be readily available in software form. This minisymposium seeks to bring together practitioners and researchers with the goal of encouraging exchange of ideas and discovering opportunities for collaboration in this direction.
Organizer: Brian C. Vermeire Imperial College London, United Kingdom
Organizer: Antony Jameson Stanford University, USA
Organizer: Peter E. Vincent Imperial College London, United Kingdom 1:30-1:50 Active Fluxes; A New HighOrder Paradigm Philip L. Roe, University of Michigan, USA 1:55-2:15 Superconvergent HDG Methods with Symmetric Stress Approximations for Stokes Flow (and Linear Elasticity) Bernardo Cockburn, University of Minnesota, Minneapolis, USA; Guosheng Fu, University of Minnesota, USA 2:20-2:40 Riemann-Solver-Free Space Time Discontinuous Galerkin Method for General Conservation Laws Shuang Z. Tu, Jackson State University, USA 2:45-3:05 Aerodynamic Simulations Using a High-Order Discontinuous Galerkin Solver Michael Brazell and Dimitri Mavriplis, University of Wyoming, USA
Organizer: Timo Betcke University College London, United Kingdom
Organizer: Andreas Kloeckner University of Illinois, USA 1:30-1:50 ExaFMM -- a Testbed for Comparing Various Implementations of the FMM Rio Yokota and David E. Keyes, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Lorena A. Barba, George Washington University, USA 1:55-2:15 Robust Implementation of Quadrature by Expansion (QBX) Manas Rachh, Courant Institute of Mathematical Sciences, New York University, USA; Leslie Greengard, Simons Foundation and Courant Institute of Mathematical Sciences, New York University, USA; Michael O’Neil, New York University, USA; Andreas Kloeckner, University of Illinois, USA
continued in next column
109
2:20-2:40 ScalFMM: A Generic Parallel Fast Multipole Library Pierre Blanchard, Berenger Bramas, and Olivier Coulaud, INRIA, France; Eric F. Darve, Stanford University, USA; Laurent Dupuy, CEA Saclay, France; Arnaud Etcheverry, INRIA, France; Guillaume Sylvand, Airbus Operation S.A.S., France 2:45-3:05 PVFMM: A Parallel Fast Multipole Method for Volume Potentials Dhairya Malhotra and George Biros, University of Texas at Austin, USA
110
Monday, March 16
MS178 Recent Advances in a Posteriori Error Estimations and Adaptive Methods Part II of II 1:30 PM-3:10 PM Room:260 B For Part 1 see MS152 Self-adaptive numerical methods provide a powerful and automatic approach in scientific computing. In particular, Adaptive Mesh Refinement (AMR) algorithms have been widely used in computational science and engineering and have become a necessary tool in computer simulations of complex natural and engineering problems. The key ingredient for success of self-adaptive numerical methods is a posteriori error estimates that are able to accurately locate sources of global and local error in the current approximation. Talks in this minisymposium will cover some recent advances in the development and analysis of both a posteriori estimators and (convergent) adaptive schemes, as well as indicate directions of future research
Organizer: Zhiqiang Cai Purdue University, USA
Organizer: Shun Zhang City University of Hong Kong, Hong Kong 1:30-1:50 A Posteriori Error Estimation in the Maximum Norm for Finite Element Methods Alan Demlow, Texas A&M University, USA 1:55-2:15 A PDE Approach to Fractional Diffusion: a Posteriori Error Analysis Abner J. Salgado, University of Tennessee, USA 2:20-2:40 Robust Residual-Based a Posteriori Error Estimation for Interface Problems: Nonconforming Elements Cuiyu He and Zhiqiang Cai, Purdue University, USA; Shun Zhang, City University of Hong Kong, Hong Kong 2:45-3:05 Robust a Priori and a Posteriori Error Estimates for Diffusion Problems with Discontinuous Coefficients Shun Zhang, City University of Hong Kong, Hong Kong
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
MS179
MS180
Large-Scale Optimization and Applications 1:30 PM-3:10 PM
Novel Spectral Approximation: Theory, Algorithms and Applications - Part II of II 1:30 PM-3:10 PM
Room:151 AB We present recent advances in the solution and modeling of large-scale optimization problem from a broad range of applications including discrete design problems, compressed sensing approaches, and more general methods for nonconvex optimization with special emphasis on scalage approaches.
Organizer: Sven Leyffer Argonne National Laboratory, USA 1:30-1:50 Augmented Lagrangian Methods for Large-Scale Nonlinear Optimization Sven Leyffer, Argonne National Laboratory, USA 1:55-2:15 Convexification Methods for Sequential Quadratic Programming Elizabeth Wong, University of California, San Diego, USA 2:20-2:40 Primal-Dual Newton Conjugate Gradients for Compressed Sensing Problems with Coherent and Redundant Dictionaries Kimon Fountoulakis, University of Edinburgh, United Kingdom 2:45-3:05 Column Generation Techniques for Large Mixed-Integer Programs Fu Lin, Argonne National Laboratory, USA
Room:151 DE For Part 1 see MS154 Spectral and high-order methods traditionally offer high-accuracy computational results in scientific computing. Yet, recent research focuses on very different and compelling advantages for applications. These include low dispersion for wave propagation problems, near-minimal degrees of freedom for oscillatory problems and great flexibility in dealing with complicated geometries. The goal of this minisymposium is to present an overview of such contemporary research topics. Methods of recent interest include Fourier extensions or continuations, radial basis functions and redundant frame discretizations. Challenges involve obtaining fast transforms and wellconditioned discretizations, or overcoming ill-conditioning in methods involving redundancy. The focus ranges from theory to applications.
Organizer: Daan Huybrechs Katholieke Universiteit Leuven, Belgium 1:30-1:50 Gaussian-Localized Polynomial Interpolation (Hermite Function Interpolation) on a Finite Interval: Are Spectrally-Accurate Rbfs Obsolete? John P. Boyd, University of Michigan, Ann Arbor, USA 1:55-2:15 A Windowed Fourier Method for Computations on the Sphere Rodrigo B. Platte, Arizona State University, USA 2:20-2:40 A Fast and WellConditioned Spectral Method Alex Townsend, University of Oxford, United Kingdom 2:45-3:05 Automatic Multivariate Approximation Tobin Driscoll, University of Delaware, USA
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
MS181
Poster Blitz
Large Scale Computing in the Geosciences Part II of II 1:30 PM-3:10 PM Room:151 G For Part 1 see MS155 Advances in scientific computing (new algorithms, open-source software, high performance computing) have made it possible to solve larger and larger problems. These advances are especially advantageous in the geosciences, where typical problems are time-dependent and three-dimensional with little or no symmetry and with important dynamics at multiple scales. Geoscientists are increasingly relying on cutting-edge software and algorithms to advance the understanding of planet Earth. This minisymposium brings together geophysicists, mathematicians, and computer scientists to discuss the latest developments at the intersection of these fields.
Organizer: Sander Rhebergen University of Oxford, United Kingdom
Organizer: Andrew J. Wathen University of Oxford, United Kingdom
Organizer: Richard F. Katz University of Oxford, United Kingdom 1:30-1:50 Multi-Scale Modelling of Granular Avalanches Anthony R. Thornton, University of Twente, Netherlands 1:55-2:15 Scalable Nonlinear Solvers for Magma Dynamics Matthew G. Knepley, University of Chicago, USA; Richard F. Katz, University of Oxford, United Kingdom 2:20-2:40 A Generic Nonlinear Solver for Geophysical Inversions Michele De Stefano, Federico Golfré Andreasi, and Alberto Secchi, Schlumberger Geosolutions, Italy 2:45-3:05 Recent Advances in Numerical Modelling of ThermoChemically Coupled Two-Phase Flow Tobias Keller, University of Oxford, United Kingdom
3:10 PM-4:30 PM Room:355
Monday Poster Session & Reception PP9 Computational Fluid Dynamics 4:30 PM-6:30 PM Room:255 A Synchronized Co-Volume Scheme for the Large-Scale Shallow Water Equations Qingshan Chen, Clemson University, USA Physics-Based Preconditioning and Dual Timestepping for Stiff Combustion Problems with Detailed Chemical Mechanisms Michael A. Hansen, University of Utah, USA Rods with Bend and Twist in a Brinkman Fluid Nguyenho Ho and Sarah D. Olson, Worcester Polytechnic Institute, USA Discrete Exterior Calculus Solution of Incompressible Flows Mamdouh S. Mohamed and Sudantha Balage, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Anil Hirani, University of Illinois at Urbana-Champaign, USA; Ravi Samtaney, King Abdullah University of Science & Technology (KAUST), Saudi Arabia Adaptive Wavelet Simulation for Weakly Compressible Flow Bounded by Solid Walls of Arbitrary Shape Naoya Okamoto, Nagoya University, Japan; Margarete Domingues, Instituto Nacional de Pesquisa Espaciais, Brazil; Katsunori Yoshimatsu, Nagoya University, Japan; Kai Schneider, Aix-Marseille Université, France
continued in next column
111
Multiple Solutions in Curved-Pipe Flow Harsh Ranjan, Indian Institute of Technology Guwahati, India Cooperation and Efficiency in Sperm Motility Patterns Owen Richfield, Paul Cripe, and Julie Simons, Tulane University, USA Efficient Simulation of Fluid-Structure Interactions Modeled by Regularized Stokes Formulation Using KernelIndependent Fast Multipole Method Minghao W. Rostami and Sarah D. Olson, Worcester Polytechnic Institute, USA Scalable Parallel Solvers for Highly Heterogeneous Nonlinear Stokes Flow Discretized with Adaptive High-Order Finite Element Johann Rudi and Toby Isaac, University of Texas at Austin, USA; Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA; Michael Gurnis, California Institute of Technology, USA; Omar Ghattas, University of Texas at Austin, USA Discrete Adjoint Openfoam and Applications Arindam Sen, Markus Towara, and Uwe Naumann, RWTH Aachen University, Germany Strategy for Efficiently Simulating Reactive Flows with Large Detailed Chemical Kinetics Hiroshi Terashima, University of Tokyo, Japan; Youhi Morii, Japan Aerospace Exploration Agency, Japan; Mitsuo Koshi, Yokohama National University, Japan Improving the Method of Regularized Stokeslets Terese Thompson, Karin Leiderman, and Hoang-Ngan Nguyen, University of California, Merced, USA An Exact and Consistent Adjoint Method for High-Fidelity Unsteady Compressible Flow Simulations Ramanathan Vishnampet, Daniel J. Bodony, and Jonathan B. Freund, University of Illinois at Urbana-Champaign, USA Periodic Stokes Flow in 2 Dimensional Space Lin Zhao, Dartmouth College, USA; Alex H. Barnett, Dartmouth College and Simons Foundation, USA
112
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
PP10 Numerical PDEs 4:30 PM-6:30 PM Room:255 Etd Spectral Deferred Correction Methods Tommaso Buvoli, University of Washington, USA Reducing the Impact of the Cfl Condition for Dispersive Wave Propagation Problems Allan P. Engsig-Karup, Technical University of Denmark, Denmark; Claes Eskilsson, Chalmers University of Technology, Sweden Spectre: An Adaptive, Parallel Discontinuous Galerkin Based Code for Relativistic Astrophysics Scott Field and Lawrence Kidder, Cornell University, USA; Jonas Lippuner and Mark Scheel, California Institute of Technology, USA; Saul Teukolsky, Cornell University, USA Explicit Strong Strong Stability Preserving Multi Step Runge-Kutta Methods Zachary J. Grant, University of Massachusetts, Dartmouth, USA Robust Residual-Based A Posteriori Error Estimate for Interface Problems: Nonconforming Linear Elements Cuiyu He and Zhiqiang Cai, Purdue University, USA; Shun Zhang, City University of Hong Kong, Hong Kong Monolithic Multi-Time-Step Coupling Methods for First and Second-Order Transient Systems Saeid Karimi and Kalyana Nakshatrala, University of Houston, USA Total Order Function Space Spectral Collocation Methods Using the Padua Points Scott Moe, University of Washington, USA Numerical Investigation of Influence of Node Alignment on Stable Calculation for Meshless Time
continued in next column
Domain Method Yoshiharu Ohi, RIKEN, Japan; Yoshihisa Fujita, Nagoya University, Japan; Taku Itoh and Soichiro Ikuno, Tokyo University of Technology, Japan Asymptotics of High-Frequency Scattering Problems Peter Opsomer and Daan Huybrechs, Katholieke Universiteit Leuven, Belgium Accurate Derivative Computation for Finite Element Codes Lee F. Ricketson, Courant Institute of Mathematical Sciences, New York University, USA; Antoine Cerfon, New York University, USA; Manas Rachh, Courant Institute of Mathematical Sciences, New York University, USA Comparison of Weak Galerkin Finite Element Method with Dgfem and Mfem Farrah Sadre-Marandi, Colorado State University, USA A Weighted Sequential Splitting Method for the 3D Maxwell’s Equations Puttha Sakkaplangkul, Oregon State University, USA A Numerical Solution to Boundary Value Problem and Volterra Integral Equations ( Linear and NonLinear) Hamid Semiyari, James Madison University, USA Adaptive Multigrid Methods for An Integrated Structural Health Monitoring (SHM) Systems for Composite Material with FluidStructure Interaction (FSI) Effect Bhuiyan Shameem M. Ebna Hai and Markus Bause, Helmut-Schmidt-Universitaet Hamburg, Germany Asymptotically Compatible Schemes for Robust Discretization of Nonlocal Models Qiang Du and Xiaochuan Tian, Columbia University, USA An Adaptive Multiscale Method for Numerical Homogenization Jack Urombo, Harare Institute of Technology, Zimbabwe
Monday, March 16
PP11 Linear Algebra and Data Analysis 4:30 PM-6:30 PM Room:255 Using Non-Authoritative Crowdsourced Data to Augment Remote Sensing Data for the Fukushima Diachii Nuclear Incident Mark Coletti and Guido Cervone, Pennsylvania State University, USA; Rebecca Goolsby, Office of Naval Research, USA The Sparse Grid Combination Technique for Solving Eigenvalue Problems Christoph Kowitz, Technische Universität München, Germany; Markus Hegland, Australian National University, Canberra, Australia; Hans-Joachim Bungartz, Technische Universität München, Germany A Tangential Interpolation Framework for MIMO Eigensystem Realization Algorithm Boris Kramer and Serkan Gugercin, Virginia Tech, USA Multi-Set Data Analysis and Simultaneous Matrix Block Diagonalization: Models and Algorithms Dana Lahat and Christian Jutten, Gipsa-Lab, France Overcoming the Gibbs Phenomenon: Fast Fourier Extension Roel Matthysen, Katholieke Universiteit Leuven, Belgium Big Graph Analytics of Human Connectome Networks Jürgen Ommen, Chih Lai, and Yulin Yang, University of St. Thomas, USA Componentwise Sensitivity of Matrix Functions and Applications Samuel Relton, University of Manchester, United Kingdom
continued on next page
2015 SIAM Conference on Computational Science and Engineering
Inducing Approximately Optimal Flow Via Truthful Mediators Ryan M. Rogers and Aaron Roth, University of Pennsylvania, USA; Jonathan Ullman, Columbia University, USA; Steven Wu, University of Pennsylvania, USA Security in Data Mining Through Cloud Computing Using Expert System Ana Sadeghitohidi and Azadeh Roozbehi, Azad Tehran University, Iran Big Data Analytics Application in Genomics Data Processing S Srinivasan, Hector Miranda, Daniel Vrinceanu, and Terence Vaughn, Texas Southern University, USA Generalized Low Rank Models Madeleine R. Udell, Corinne Horn, Reza Zadeh, and Stephen Boyd, Stanford University, USA A Structured Cholesky Factorization for Fock Matrix Construction Joseph Vokt and Edmond Chow, Georgia Institute of Technology, USA Parallel Bayesian Global Optimization, With Application To Metrics Optimization at Yelp Jialei Wang and Peter I. Frazier, Cornell University, USA; Scott Clark and Eric Liu, Yelp Inc., USA
113
Monday, March 16
Monday, March 16
PP12
PP13
Biomedical Applications 4:30 PM-6:30 PM
Statistical Methods and Uncertainty Quantification 4:30 PM-6:30 PM
Room:255 The Modified Bidomain Model with Periodic Diffusive Inclusions Andjela Davidovic, INRIA Bordeaux SudOuest, France; Yves Coudiere, Université Bordeaux I, France; Clair Poignard, INRIA Bordeaux Sud-Ouest, France
Room:255
Simulation-Based Solute Transport in Kidney Cells Monica Nadal-Quiros and Aniel NievesGonzalez, University of Puerto Rico, Puerto Rico; Leon Moore, State University of New York Health Science Center at Brooklyn, USA; Mariano Marcano, University of Puerto Rico, Puerto Rico
The Combined Block by Block Monte Carlo Methods for Numerical Treatment of the Mixed Nonlinear Stochastic Integral Equation Abdallah A. Badr, University of Business and Technology, Jeddah, Saudi Arabia
Magnetic Nanoparticle Stochastic Dynamics for Biomedical Applications Daniel B. Reeves and John Weaver, Dartmouth College, USA An Adaptive Markov Chain Monte Carlo Method Applied to Simulation of a Tumor Growth Model Qing Wang and Zhijun Wang, Shepherd University, USA; David Klinke, West Virginia University, USA “Allostery”: A Python Package for Network Analyses of Biomolecular Simulations Yuhang Wang and Emad Tajkhorshid, University of Illinois at UrbanaChampaign, USA Numerical Simulation of a Tumor Cell Population Growth Dynamics Model Using Genetic Algorithm Zhijun Wang and Qing Wang, Shepherd University, USA Numerical Methods for Protein Adsorption in Porous Membranes Anastasia B. Wilson, Clemson University, USA Modeling Core Body Temperature during Exercise Yeonjoo Yoo, Indiana University - Purdue University Indianapolis, USA
Decomposition-Based Uncertainty Quantification with Application to Environmental Impacts of Aviation Sergio Amaral, Massachusetts Institute of Technology, USA
Computational Investigation of Quasi-Random Sequences for Error Estimation Hongmei Chi, Florida A & M University, USA Optimal Source Encoding in Medium Parameter Reconstruction Problems Benjamin Crestel and Omar Ghattas, University of Texas at Austin, USA; Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA Randomized Likelihood Method: A Scalable Approach to Big Data in Large-Scale Pde-Constrained Bayesian Inverse Problems Aaron Myers, Tan Bui-Thanh, and Ellen Le, University of Texas at Austin, USA MUQ (MIT Uncertainty Quantification): Flexible Software for Connecting Algorithms and Applications Matthew Parno, Patrick R. Conrad, Andrew Davis, and Youssef M. Marzouk, Massachusetts Institute of Technology, USA A Stochastic Dynamic Programming Method for Controlling a Combined Hydro/Wind Power Producer Kyle Perline, Cornell University, USA
Ensemble Kalman Filters for Dynamic Dipole Estimation from Magnetoencephalography Lijun Yu, Daniela Calvetti, and Erkki Somersalo, Case Western Reserve University, USA continued on next page
114
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
PP13 Statistical Methods and Uncertainty Quantification 4:30 PM-6:30 PM continued
Efficient Error Estimation for Elliptic PDEs with Random Data Catherine Powell, University of Manchester, United Kingdom; Alex Bespalov, University of Birmingham, United Kingdom; David Silvester, University of Manchester, United Kingdom Inference of Constitutive Parameters in a Nonlinear Stokes Mantle Flow Model Vishagan Ratnaswamy, California Institute of Technology, USA; Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA; Michael Gurnis, California Institute of Technology, USA; Omar Ghattas, University of Texas at Austin, USA Uncertainty Quantification for Thermally Driven Flow Sebastian Ullmann and Lang Jens, TU Darmstadt, Germany Computational and Statistical Tradeoffs: a Framework Alexander Volfovsky, Edoardo Airoldi, and Daniel Sussman, Harvard University, USA Efficiency of the Girsanov Transformation Approach for Parametric Sensitivity Analysis of Stochastic Chemical Kinetics Ting Wang and Muruhan Rathinam, University of Maryland, Baltimore County, USA Convergence of the Robbins-Monro Algorithm in Infinite Dimensional Hilbert Spaces Daniel Watkins and Gideon Simpson, Drexel University, USA Utilizing Adjoint-Based Techniques to Effectively Perform Uq on Discontinuous Responses Tim Wildey, Eric C. Cyr, and John Shadid, Sandia National Laboratories, USA
continued in next column
Optimization of Modeled Land Surface Fluxes by Bayesian Parameter Calibration Tony E. Wong, David Noone, and William Kleiber, University of Colorado Boulder, USA Monte Carlo Sampling of the Aggregation of Lattice Biopolymers in Membrane Yuanwei Xu and Mark Rodger, University of Warwick, United Kingdom
Monday, March 16
PP14 CSE Methods 4:30 PM-6:30 PM Room:255 Visualising Protein Sequence Alignment Shaimaa M. Aljuhani, Al-Imam Muhammad bin Saud University, Saudi Arabia; Prof. Teresa Attwood, University of Manchester, United Kingdom; Dr. Tony Shardlow, University of Bath, United Kingdom Reconstructing Physically Realistic Flow Fields from Sparse Experimental Data Iliass Azijli, Richard Dwight, Jan Schneiders, and Hester Bijl, Delft University of Technology, Netherlands Parallel-in-Time Integration with Pfasst++ Torbjörn Klatt and Robert Speck, Jülich Supercomputing Centre, Germany; Mathias Winkel and Daniel Ruprecht, Universita’ della Svizzera Italiana, Italy; Matthew Emmett, Lawrence Berkeley National Laboratory, USA An Optimization-Based Approach Toward Elastoplasticity: Introducing a Projected Newton Algorithm Zahra S. Lotfian and Mettupalayam Sivaselvan, State University of New York at Buffalo, USA Redesigning Laser-Plasma Simulations to Optimize the Use of Limited Memory Bandwidth Eileen R. Martin, Stanford University, USA; Steve Langer, Lawrence Livermore National Laboratory, USA Parallel Numerics for Partitioned Multiphysics Coupling Miriam Mehl, Universität Stuttgart, Germany; Benjamin Uekermann, Technische Universität München, Germany; Florian Lindner, Universität Stuttgart, Germany Applied Math and Cs R&D on Doe Leadership Computing Facilities Paul C. Messina, Argonne National Laboratory, USA
continued on next page
2015 SIAM Conference on Computational Science and Engineering
Simple Yet Fast Integration Method Using Qss for Stiff Chemical Kinetic Odes Youhi Morii, Japan Aerospace Exploration Agency, Japan; Hiroshi Terashima, University of Tokyo, Japan; Mitsuo Koshi, Yokohama National University, Japan; Taro Shimizu and Eiji Shima, Japan Aerospace Exploration Agency, Japan PtychoLib: Parallel Ptychographic Reconstruction Youssef Nashed, David Vine, and Tom Peterka, Argonne National Laboratory, USA; Junjing Deng, Northwestern University, USA; Rob Ross and Chris Jacobsen, Argonne National Laboratory, USA Efficient Numerical Algorithm for Virtual Design in Nanoplasmonics Alexandra Ortan, University of Minnesota, USA Model-Reduction for Closed-Loop Control of Unsteady Flows Using Plasma Actuators Laura Pasquale, Paul Houston, and Pericle Zanchetta, University of Nottingham, United Kingdom Moving Pictures: Animating Still Images Michael Pilosov, University of Colorado, Denver, USA Bridging Multiple Structural Scales with a Generalized Finite Element Method Julia A. Plews and C. Armando Duarte, University of Illinois at UrbanaChampaign, USA A Parallelization Strategy for LargeScale Vibronic Coupling Calculations Scott Rabidoux, Victor Eijkhout, and John Stanton, University of Texas at Austin, USA Some Numerical Methods for Modified Bessel Functions Juri M. Rappoport, Russian Academy of Sciences, Russia Oof: An Object-Oriented FiniteElement Solver for Materials Science Andrew Reid and Stephen Langer, National Institute of Standards and Technology, USA
continued in next column
Analyzing and Classifying “TwoCycles” of Trigonometric Functions in Newton’s Method Morgan Rupard and Jennifer Switkes, California Polytechnic State University, Pomona, USA Orbital Localization in Madness Bryan E. Sundahl, Stony Brook University, USA; Robert Harrison, Brookhaven National Laboratory, USA; Scott Thornton, Stony Brook University, USA Assessing Artifacts: Segmentation of Damaged Images Nick Takaki, Carnegie Mellon University, USA Analysis of Anderson Acceleration for Coupled Neutronic and Thermal Hydraulic Calculations in a Light Water Reactor Alexander R. Toth and C.T. Kelley, North Carolina State University, USA; Stuart Slattery, Steven Hamilton, and Kevin Clarno, Oak Ridge National Laboratory, USA; Roger Pawlowski, Sandia National Laboratories, USA The Rapid Optimization Library (rol) in Trilinos Bart G. Van Bloemen Waanders, Drew P. Kouri, and Denis Ridzal, Sandia National Laboratories, USA Integrating Software Tools to Parallel Adaptive Simulations of Fusion Plasma in Tokamaks Fan Zhang, Mark S. Shephard, and E. Seegyoung Seol, Rensselaer Polytechnic Institute, USA
115
Monday, March 16
PP15 CSE Applications 4:30 PM-6:30 PM Room:255 Using Radar Imagery Data to Invert for Maritime Environments Vasileios Fountoulakis and Christopher J. Earls, Cornell University, USA Thermal Imaging of Sub-Pixel Cracks Through Metal Plates Andrew Loeb and Christopher Earls, Cornell University, USA Numerical Simulation of Ni Grain Growth in a Thermal Gradient John A. Mitchell and Veena Tikare, Sandia National Laboratories, USA Supply Chain Disruptions Thomas Morrisey and Ravi Prasad, Infosys, USA Computational and Experimental Analysis of Dental Implants under Different Loading Conditions and Locations Emre Ozyilmaz, Eda Ozyilmaz, and Halil Aykul, Hitit University, Turkey; Mehmet Dalkiz, Mustafa Kemal University, Turkey; Ahmet Cini, Hitit University, Turkey Math Projects with Tracker Video Analysis Euguenia Peterson, Richard J. Daley College, USA Investigation of Numerical Models for New High Temperature Superconductors Chad Sockwell, Florida State University, USA Characterization of Ternary Eutectic Solidification Patterns from PhaseField Simulations and Experimental Micrographs Philipp Steinmetz, Karlsruhe Institute of Technology, Germany; Johannes Hötzer, Hochschule Karlsruhe Technik und Wirtschaft, Germany; Marcus Jainta and Britta Nestler, Karlsruhe Institute of Technology, Germany; Yuksel Yabansu and Surya Kalidindi, Georgia Institute of Technology, USA
continued on next page
116
Monday, March 16
PP15 CSE Applications 4:30 PM-6:30 PM continued
Numerical Realization of An Open Pit Mine Planning Problem Nikolai Strogies and Andreas Griewank, Humboldt University Berlin, Germany Multidisciplinary Development of An Autonomous Underwater Vehicle: Cooperative Fleet for Surveillance Mission Ci Wen, Stacey Joseph-Ellison, Junzhen Shao, Qi Zhou, Jonathan Jaworski, and Zakaria Daud, Embry-Riddle Aeronautical University, USA Towards Real-Time Blob-Filaments Detection in Fusion Plasma Lingfei Wu, College of William & Mary, USA; Kesheng Wu and Alex Sim, Lawrence Berkeley National Laboratory, USA; Andreas Stathopoulos, College of William & Mary, USA
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
PP201
PP202
Minisymposterium -A Mathematical Environment for Quantifying Uncertainty: Integrated and Optimized at the eXtreme scale (EQUINOX) 4:30 PM-6:30 PM
Minisymposterium -Advanced Discretizations for Complex Applications 4:30 PM-6:30 PM Room:255
Room:255
Organizers: Gianmarco Manzini, Los Alamos National Laboratory, USA
Organizer: Clayton G. Webster, Oak Ridge National Laboratory, USA
Konstantin Lipnikov, Los Alamos National Laboratory, USA
Forward Backward Doubly Stochastic Differential Equations and Applications to The Optimal Filtering Problem Feng Bao, Oak Ridge National Laboratory, USA; Yanzhao Cao, Auburn University, USA; Clayton G. Webster and Guannan Zhang, Oak Ridge National Laboratory, USA Embedded Sampling-Based Uncertainty Quantification Approaches for Emerging Computer Architectures Eric Phipps, Marta D’Elia, H. Carter Edwards, Jonathan J. Hu, and Siva Rajamanickam, Sandia National Laboratories, USA A Unified Framework for Uncertainty and Sensitivity Analysis of Computational Models with Many Input Parameters Li Gu and C. F. Jeff Wu, Georgia Institute of Technology, USA Florida State University Efforts Withing the Equinox Project Max Gunzburger, Florida State University, USA A Mathematical Environment for Quantifying Uncertainty: Integrated and Optimized at the EXtreme Scale (equinox) Clayton G. Webster, Oak Ridge National Laboratory, USA Hierarchical Acceleration of Multilevel Methods for Pdes with Random Input Data Guannan Zhang, Oak Ridge National Laboratory, USA
Compatible Discrete Operator Schemes for Advection-Diffusion Equations Pierre Cantin and Alexandre Ern, Université Paris-Est, France; Jérome Bonelle, EDF, France Upwinding in the Mimetic Finite Difference Method for Richards’ Equation Konstantin Lipnikov, Los Alamos National Laboratory, USA The Virtual Element Method Gianmarco Manzini, Los Alamos National Laboratory, USA Virtual Element Methods for Convection-Diffusion Problems Oliver Sutton and Andrea Cangiani, University of Leicester, United Kingdom; Gianmarco Manzini, Los Alamos National Laboratory, USA; Alessandro Russo, Milano University, Italy
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
PP203
PP204
Minisymposterium -Advances in Numerical Methods for Interface Problems 4:30 PM-6:30 PM
Minisymposterium -CSE Software 4:30 PM-6:30 PM Room:255
Room:255
Organizers: Anders Logg, Chalmers University of Technology, Sweden
Organizers; Pavel Bochev, Sandia National Laboratories, USA
Chebfun Anthony Austin, University of Oxford, United Kingdom
Christopher Siefert, Sandia National Laboratories, USA Optimal Control for Mass Conservative Level Set Methods Christopher Basting and Dmitri Kuzmin, Technische Universität Dortmund, Germany A New Partitioned Algorithm for Explicit Elastodynamics Based on Variational Flux Recovery Pavel Bochev and Paul Kuberry, Sandia National Laboratories, USA Analysis of a Fluid-Structure Interaction Problem Decoupled by Optimal Control Paul A. Kuberry and Hyesuk Lee, Clemson University, USA Feature-Preserving Finite Element Transport Across Interfaces: Part 1, Optimization-Based Transport Denis Ridzal, Kara Peterson, and Pavel Bochev, Sandia National Laboratories, USA Feature-Preserving Finite Element Transport Across Interfaces: Part 2, Direct Flux Recovery Kara Peterson, Pavel Bochev, and Denis Ridzal, Sandia National Laboratories, USA Higher Order Finite Element Methods for Interface Problems Manuel A. Sanchez-Uribe, Brown University, USA Extended and Conformal Decomposition Finite Elements for 3D Compatible Discretizations Christopher Siefert, Richard Kramer, Pavel Bochev, and Thomas Voth, Sandia National Laboratories, USA
Feel++: A Versatile High Performance Finite Element Embedded Library into C++ Vincent Chabannes, Université de Grenoble I, France Dolfin-Adjoint Simon W. Funke and Marie E. Rognes, Simula Research Laboratory, Norway; Patrick E. Farrell, University of Oxford, United Kingdom; David Ham, Imperial College London, United Kingdom Building Performance Transportable Codes for Extreme Scale William D. Gropp, University of Illinois at Urbana-Champaign, USA Firedrake: Automating Finite Element by Composing Abstractions David Ham, Florian Rathgeber, Lawrence Mitchell, Michael Lange, Andrew McRae, Gheorghe-Teodor Bercea, Fabio Luporini, and Paul Kelly, Imperial College London, United Kingdom The DEAL.II Finite Element Library Timo Heister, Clemson University, USA; Wolfgang Bangerth, Texas A&M University, USA; Guido Kanschat and Matthias Maier, Universität Heidelberg, Germany An Overview of the Trilinos Project Michael Heroux, Sandia National Laboratories, USA FEniCS High Performance Computing with Applications in Aerodynamics, Environmental Science and Biomedicine Johan Hoffman, Royal Institute of Technology, Stockholm, Sweden EMatter: A Materials Simulation Framework As a Service Dmitry A. Karpeyev, University of Chicago, USA
continued in next column
117
Composability in Petsc Matthew G. Knepley, University of Chicago, USA; Jed Brown, Argonne National Laboratory, USA and University of Colorado Boulder, USA Fenics Anders Logg, Chalmers University of Technology, Sweden Sigma: Scalable Interface for Geometry and Mesh Based Applications Vijay Mahadevan, Iulian Grindeanu, Rajeev Jain, Navamita Ray, and Danqing Wu, Argonne National Laboratory, USA; Paul Wilson, University of Wisconsin, Madison, USA Dune - The Distributed and Unified Numerics Environment Steffen Müthing, Heidelberg University, Germany Elemental Jack L. Poulson, Stanford University, USA Jupyter Widgets: Interactive Computing Through the Browser in Any Programming Language Min Ragan-Kelley and Fernando Perez, University of California, Berkeley, USA Camellia: A Software Framework for Discontinuous Petrov-Galerkin Methods Nathan Roberts, Argonne National Laboratory, USA ViennaCL - Fast Linear Algebra for Multi and Many-Core Architectures Karl Rupp, Philippe Tillet, Toby St Clere Smithe, Namik Karovic, Josef Weinbub, and Florian Rudolf, Vienna University of Technology, Austria
118
Monday, March 16
PP205 Minisymposterium -High Performance Linear Algebra Operations on GPUs 4:30 PM-6:30 PM Room:255 Organizer: Luke Olson, University of Illinois at Urbana-Champaign, USA Radical Optimization Techniques for Asynchronous Iterative Algorithms on Gpus Hartwig Anzt, University of Tennessee, USA; Edmond Chow, Georgia Institute of Technology, USA; Jack J. Dongarra, University of Tennessee, Knoxville, USA Experiences in Autotuning Linear Algebra Operations for Energy Minimization on Gpus Hartwig Anzt and Blake Haugen, University of Tennessee, USA; Jakub Kurzak and Jack J. Dongarra, University of Tennessee, Knoxville, USA
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
Monday, March 16
PP206
PP207
Minisymposterium -Massively Parallel Multiphysics Simulations Within the Uintah Computational Framework 4:30 PM-6:30 PM
Minisymposterium -- Mixed Finite Element Methods 4:30 PM-6:30 PM
Room:255
The Periodic Table of the Finite Elements Douglas N. Arnold, University of Minnesota, USA; Anders Logg, Chalmers University of Technology, Sweden
Organizer: James C. Sutherland, University of Utah, USA Parallel Petascale Modeling of Transportation Accidents Involving High Explosives Jacqueline Beckvermit, Andrew Bezdjian, Todd Harman, John A. Schmidt, and Martin Berzins, University of Utah, USA; Chuck Wight, Weber State University, USA Using Uintah:mpmice for High Resolution Urban Flow Studies Arash Nemati Hayati, Rob Stoll, Todd Harman, and Eric Pardyjak, University of Utah, USA
CUSP: A Parallel Sparse Matrix Package for Gpus Steven Dalton and Luke Olson, University of Illinois at Urbana-Champaign, USA
Radiation Modeling Using Reverse Monte Carlo Ray Tracing Within the Uintah Framework Alan Humphrey, University of Utah, USA
Sparse Matrix-Matrix Multiplication on High-Throughput Architectures Luke Olson and Steven Dalton, University of Illinois at Urbana-Champaign, USA
Wasatch: A CPU/GPU-Ready Multiphysics Code Using a Domain Specific Language Tony Saad, Abhishek Bagusetty, and James C. Sutherland, University of Utah, USA Applied Large Eddy Simulation: Validation and Uncertainty Quantification of Lab and Pilot-Scale, Oxy-Coal Boiler Simulations Jeremy Thornock and Wu Yuxin, University of Utah, USA; Ben Isaac, University of Utah, USA & Université Libre de Bruxelles, Belgium; Sean Smith and Philip J. Smith, University of Utah, USA
Room:255 Organizer: Douglas N. Arnold, University of Minnesota, USA
Convolution-Translation and Bounded Cochain Projections for the Elasticity Complex Gerard Awanou, Northern Illinois University, USA What is a Good Linear Finite Element... On a Generic Polytope? Andrew Gillette, University of Arizona, USA; Alexander Rand, CD-adapco, USA Superconvergence of Nonconforming Fems Based Mixed Fems Jun Hu, Peking University, China Stokes Elements on Cubic Meshes Yielding Divergence-Free Approximations Michael J. Neilan and Duygu Sap, University of Pittsburgh, USA Weak Galerkin Finite Element Methods Junping Wang, National Science Foundation, USA; Chunmei Wang, Georgia Institute of Technology, USA; Xiu Ye, University of Arkansas at Little Rock, USA
2015 SIAM Conference on Computational Science and Engineering
Monday, March 16
PP208 Minisymposterium -- Newest Bayesian Approaches to Modeling Complex Systems 4:30 PM-6:30 PM Room:255 Organizers: Alberto Giovanni Busetto, University of California, Santa Barbara, USA James Beck, California Institute of Technology, USA Approximate Active Bayesian Inference of Nonlinear Dynamical Systems Alberto Giovanni Busetto, University of California, Santa Barbara, USA Variational Reformulation of Bayesian Inverse Problems Ilias Bilionis, Purdue University, USA; Panagiotis Tsilifis, University of Southern California, USA; Nicholas Zabaras, Cornell University, USA Bayesian Model Selection for Exploring Mechanisms Contributing to Differential Signaling Pencho Yordanov and Joerg Stelling, ETH Zürich, Switzerland
119
Tuesday, March 17
Tuesday, March 17
Registration 7:45 AM-5:00 PM
Featured Minisymposium: CSE Software 10:00 AM-11:40 AM
Room:East Foyer
MS182
Room:355 A
Announcements 8:10 AM-8:15 AM Room:355
IP7 A Calculus for the Optimal Quantification of Uncertainties 8:15 AM-9:00 AM Room:355 Chair: Susan E. Minkoff, University of Texas at Dallas, USA The past century has seen a steady increase in the need of estimating and predicting complex systems and making (possibly critical) decisions with limited information. With this purpose, this talk will describe the development of a form of calculus allowing for the (computational) manipulation of infinite dimensional information structures and its application to the optimal quantification of uncertainties in complex systems and the scientific computation of optimal statistical estimators/models. Specific examples will be discussed to illustrate how this form of calculus could also be used to facilitate/guide the process of scientific discovery.
Houman Owhadi California Institute of Technology, USA
Prize Award Ceremony 9:00 AM-9:30 AM Room:355
Coffee Break 9:30 AM-10:00 AM Room:255
This featured minisymposium reflects recent progress with tools and techniques for developing, adapting, using, and testing CSE software. The first talk addresses an eco system of computational tools for increased efficiency and reliability of scientific workflows. The next talk presents a software framework in C++, MOOSE, which simplifies building advanced multiphysics models. Another talk explains how domain specific languages for PDEs in Python allow automatic generation of adjoint models in C++ with high performance. Finally, the recent Julia language is presented, which combines the expressiveness and convenience of MATLAB/Python with the speed of Fortran/C.
Organizer: Hans Petter Langtangen Simula Research Laboratory and University of Oslo, Norway 10:00-10:20 Moose: An Open Source Platform For Rapid Development of Multiphysics Simulation Tools Derek R. Gaston, Cody Permann, David Andrs, John Peterson, and Andrew Slaughter, Idaho National Laboratory, USA 10:25-10:45 Domain Specific Languages and Automated Code Generation: High Expressiveness and High Performance Patrick E. Farrell, University of Oxford, United Kingdom 10:50-11:10 Julia - a Fast New Language for Technical Computing Stephan Karpinski, Massachusetts Institute of Technology, USA 11:15-11:35 Project Jupyter: a Language-Independent Architecture for Cse, from Interactive Computing to Reproducible Publications Fernando Perez, University of California, Berkeley, USA
120
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS183 Challenges in Optimization with Uncertainty and Noisy Data 10:00 AM-11:40 AM Room:355 D In many practical applications of optimization, the objective function and constraints are not known precisely. This lack of knowledge may stem from noisy or inaccurate evaluations of the objective and constraints, or from uncertain parameters in the underlying models. A probabilistic approach to the latter yields problems of robust optimization, encompassing a variety of operators and risk measures. Yet the two cases are linked, in that evaluation of a probabilistic risk measure is itself subject to inevitable and often significant numerical errors. This minisymposium presents recent algorithmic developments for optimization, aimed at efficient and reliable solutions of problems with uncertainty.
Organizer: Florian Augustin Massachusetts Institute of Technology, USA
Organizer: Youssef M. Marzouk Massachusetts Institute of Technology, USA 10:00-10:20 Bayesian Global Optimization of Expensive Functions with Low-Dimensional Noise Jing Xie, Cornell University, USA; Sethuraman Sankaran, HeartFlow, Inc, USA; Abhay Ramachandra, University of California, San Diego, USA; Saleh Elmohamed, Cornell University, USA; Alison Marsden, University of California, San Diego, USA; Peter I. Frazier, Cornell University, USA 10:25-10:45 Modeling An Augmented Lagrangian for Improved Blackbox Constrained Optimization Robert Gramacy, University of Chicago, USA; Genetha Gray, Sandia National Laboratories, USA; Sebastien Le Digabel, École Polytechnique de Montréal, Canada; Herbert Lee, University of California, Santa Cruz, USA; Pritam Ranjan, Acadia University, Canada; Garth Wells, University of Cambridge, United Kingdom; Stefan Wild, Argonne National Laboratory, USA
continued in next column
10:50-11:10 A Model and Variance Reduction Method for Computing Statistical Outputs of Stochastic Partial Differential Equations Ferran Vidal-Codina, Cuong Nguyen, and Jaime Peraire, Massachusetts Institute of Technology, USA; Michael B. Giles, University of Oxford, United Kingdom 11:15-11:35 Topology Optimization of a Permanent-Magnet Synchronous Machine under Uncertainties Roland Pulch, University of Greifswald, Germany; Piotr Putek, Bergische Universität Wuppertal, Germany
Tuesday, March 17
MS184 First Principle Methods and Applications for Computational Materials Science and Chemistry Part I of III 10:00 AM-11:40 AM Room:250 A For Part 2 see MS209 First principle electronic structure calculations are among the most challenging and computationally demanding science and engineering problems. This minisymposium aims at presenting and discussing new numerical methods for achieving ever higher level of accuracy and efficiency in ground and excited density functional theory calculations, and their applications in computational materials science and chemistry. We propose to bring together physicists, chemists and materials scientists who are involved in improving the numerical development of widely known quantum chemistry and solid-state physics application software packages, with mathematicians and computer scientists who are focusing on advancing the required state-of-theart mathematical algorithms and parallel implementation.
Organizer: Lin Lin University of California, Berkeley, USA
Organizer: Eric Polizzi University of Massachusetts, Amherst, USA
Organizer: Chao Yang Lawrence Berkeley National Laboratory, USA 10:00-10:20 Accurate All-Electron Electronic Structure Theory for Large Systems Volker Blum, Duke University, USA 10:25-10:45 Enabling Large-Scale Hybrid Density Functional Theory Calculations Robert A. DiStasio, Jr., Princeton University, USA
continued on next page
2015 SIAM Conference on Computational Science and Engineering
10:50-11:10 Finite Elements for Large, Accurate Quantum Mechanical Materials Calculations: from Classical to Enriched to Discontinuous John Pask, Lawrence Livermore National Laboratory, USA 11:15-11:35 Enabling Large Scale LAPW DFT Calculations by a Scalable Iterative Eigensolver Daniel Wortmann, Forschungszentrum Jülich, Germany; Edoardo A. Di Napoli, Jülich Supercomputing Centre, Germany; Mario Berljafa, University of Manchester, United Kingdom
Tuesday, March 17
MS185 High-dimensional Approximation and Integration: Analysis and Computation - Part V of V 10:00 AM-11:40 AM Room:250 B For Part 4 see MS159 This minisymposium focuses on the fundamental problem of how to accurately approximate solutions of both forward and inverse high-dimensional functions. Predicting the behavior of complex phenomena relies on constructing solutions in terms of high dimensional spaces, particularly in the case when the input data (coefficients, forcing terms, initial and boundary conditions, geometry) are affected by large amounts of uncertainty. The resulting explosion in computational effort is a symptom of the curse of dimensionality and this symposium aims at exploring breakthroughs in sparse sampling and representations, nonlinear and greedy approximations, compressed sensing, multilevel methods, and ”best-N-term” approximations.
Organizer: Guannan Zhang Oak Ridge National Laboratory, USA
Organizer: Albert Cohen Université Pierre et Marie Curie, France
Organizer: Max Gunzburger Florida State University, USA
Organizer: Clayton G. Webster Oak Ridge National Laboratory, USA 10:00-10:20 Solving Optimal Feedback Control Problems for Partial Differential Equations Using Adaptive Sparse Grids Jochen Garcke, University of Bonn, Germany; Axel Kroner, RICAM, Austrian Academy of Sciences, Austria 10:25-10:45 Dimension-Independent, Likelihood-Informed Mcmc Sampling Algorithms for Bayesian Inverse Problems Kody Law, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Tiangang Cui and Youssef M. Marzouk, Massachusetts Institute of Technology, USA
continued in next column
121
10:50-11:10 Data and Uncertainties: Representation of High-Dimensional Dependencies Using Adaptive Sparse Grids Dirk Pfluger and Fabian Franzelin, University of Stuttgart, Germany 11:15-11:35 Numerical Solution of Elliptic Diffusion Problems on Random Domains Michael Peters and Helmut Harbrecht, Universität Basel, Switzerland
122
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS186 Resilient, Extreme-scale Numerical Algorithms Part I of II 10:00 AM-11:40 AM Room:250 C For Part 2 see MS210 Computer architectures are undergoing fundamental changes that will require the rethinking and invention of numerical algorithms suitable for computing at extreme scales (over 100 PFlops). Numerical solver technologies are needed that favor computation over data motion, that expose much greater concurrency, and that are resilient to increasing soft and hard faults. This minisymposium will feature progress in the development of such algorithms by several projects funded by the DOE Resilient Extreme-Scale Solvers initiative. A breadth of topics will be addressed, including algebraic multigrid, methods for hyperbolic conservation laws, Particle-InCell algorithms, and stochastic methods for linear systems.
Organizer: John Loffeld Lawrence Livermore National Laboratory, USA
Organizer: Andrew Myers Lawrence Berkeley National Laboratory, USA
Organizer: Jeffrey A. Hittinger Lawrence Livermore National Laboratory, USA 10:00-10:20 Attaining High Arithmetic Intensity in Finite-volume Methods through High-order Quadratures John Loffeld and Jeffrey A. Hittinger, Lawrence Livermore National Laboratory, USA 10:25-10:45 Parallel Spectral ElementBased Agglomeration Algebraic Multigrid for Porous Media Flow Andrew T. Barker, Lawrence Livermore National Laboratory, USA; Delyan Kalchev, University of Colorado, USA; Panayot Vassilevski and Umberto E. Villa, Lawrence Livermore National Laboratory, USA
continued in next column
10:50-11:10 Incorporating Error Detection and Recovery into Hierarchically Semi-Separable Matrix Operations Brian Austin, Alex Druinsky, Xiaoye Sherry Li, Osni A. Marques, and Eric Roman, Lawrence Berkeley National Laboratory, USA
Tuesday, March 17
11:15-11:35 Controlling Numerical Error in Particle-In-Cell Simulations of Collisionless Dark Matter Andrew Myers, Brian Van Straalen, and Colella Phillip, Lawrence Berkeley National Laboratory, USA
Room:250 D
MS187 Recent Advances in Model Reduction - Part I of V 10:00 AM-11:40 AM For Part 2 see MS211 Model reduction has become an increasingly important tool to mitigate the computational burden of modeling and simulation in timecritical (e.g., model-predictive control) and many query (e.g., Bayesian inversion) applications. This minisymposium presents recent advances that address the primary challenges facing such methods, such as preserving intrinsic problem structure, handling high-dimensional parameter spaces, integration with existing simulation codes, stability, optimization and uncertainty quantification and control.
Organizer: Kevin T. Carlberg Sandia National Laboratories, USA
Organizer: Gianluigi Rozza SISSA, International School for Advanced Studies, Trieste, Italy 10:00-10:20 Recent Advances in Reduced Order Modelling in Computational Fluid Dynamics within EU-MORNET COST Activities Gianluigi Rozza and Giuseppe Pitton, SISSA, International School for Advanced Studies, Trieste, Italy; Annalisa Quaini, University of Houston, USA 10:25-10:45 A Minimum-residual Mixed Reduced Basis Method: Exact Residual Certification and Simultaneous Finite-element Reduced-basis Adaptive Refinement Masayuki Yano, Massachusetts Institute of Technology, USA 10:50-11:10 Reduced Basis Method for Uncertainty Quantification Problems: A Recent Update Peng Chen, ETH Zürich, Switzerland; Gianluigi Rozza, SISSA, International School for Advanced Studies, Trieste, Italy 11:15-11:35 Accelerating the Solution of Inverse Problems Using ReducedOrder Models David Amsallem, Stanford University, USA
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
Tuesday, March 17
MS188
MS189
Advances in Time-stepping Methods - Part III of V 10:00 AM-11:40 AM
Sparse and Compressible Representations: Theory, Algorithms and Applications - Part I of III 10:00 AM-11:40 AM
Room:250 E For Part 2 see MS162 For Part 4 see MS212 Time-stepping methods are numerical methods for the time evolution of ODEs, DAEs, and PDEs in the method-of-lines framework. Many applications require specialized time-stepping methods in order to enhance efficiency or capture particular theoretical properties such as positivity, symplecticity, and strong-stabilitypreservation or practical properties such as parallelization on emerging high-performance architectures, component (e.g., multirate or local time-stepping) and physics (IMEX or semi-implicit) splittings based on sub-system properties, and reliable error estimation and control. The talks in this minisymposium will describe recent novel developments in time-stepping methods and demonstrate their benefits in practical applications.
Room:250 F
Organizer: Michael Guenther
For Part 2 see MS213 By focusing effort on identifying dominant terms of a function representation one can often accurately approximate the function from limited data. Sparsity detection also referred to as compressed sensing and l1-minimization is one approach that is extremely effective at identifying the dominant terms of a function expansion. Recently there have been many advances in l1-minimization including strategies for sampling, preconditioning and highdimensional basis selection that attempt to maximize the accuracy of recovered signals whilst minimizing the number of signal samples. This minisymposium presents recent theoretical results, numerical methods and applications of sparse approximations, including but not limited to uncertainty quantification (UQ) and sensitivity analysis (SA).
Bergische Universität, Germany
Organizer: John D. Jakeman
Organizer: Adrian Sandu
Sandia National Laboratories, USA
Virginia Tech, USA
Organizer: Khachik Sargsyan
Organizer: Raymond J. Spiteri
Sandia National Laboratories, USA
Organizer: Emil M. Constantinescu Argonne National Laboratory, USA
University of Saskatchewan, Canada 10:00-10:20 A New Look at Global Error Estimation in Differential Equations Emil M. Constantinescu, Argonne National Laboratory, USA 10:25-10:45 Reliable Adaptive Time Stepping for Systems of ODEs, DDEs and BVPs Wayne Enright, University of Toronto, Canada 10:50-11:10 High Order Semi-implicit Schemes for Evolutionary Non Linear Partial Differential Equations and Applications Sebastiano Boscarino, University of Catania, Italy
10:00-10:20 Sparse Polynomial Chaos Expansions for Time-Dependent Problems Bruno Sudret and Chu Mai, ETH Zürich, Switzerland 10:25-10:45 Adaptive Compressive Sensing Method for Uncertainty Quantification Xiu Yang, Pacific Northwest National Laboratory, USA; Xiaoliang Wan, Louisiana State University, USA; Huan Lei and Guang Lin, Pacific Northwest National Laboratory, USA; George E. Karniadakis, Brown University, USA
11:15-11:35 On the Construction of Robust Additive Runge-Kutta Methods Inmaculada Higueras, Universidad Pública de Navarra, Spain continued in next column
123
10:50-11:10 Stochastic Collocation Methods Via L1 Minimization Tao Zhou, Chinese Academy of Sciences, China; Ling Guo, Shanghai Normal University, China; Dongbin Xiu, Purdue University, USA 11:15-11:35 A Sparse Multiresolution Regression Framework for Uncertainty Quantification Daniele E. Schiavazzi, University of California, San Diego, USA; Alireza Doostan, University of Colorado Boulder, USA; Gianluca Iaccarino, Stanford University, USA
124
Tuesday, March 17
MS190 The Use of Adjoints in Earth System Modeling/Sciences 10:00 AM-11:40 AM Room:251 A Derivatives can be used in many contexts including optimization, sensitivity analysis, and uncertainty quantification. Earth system models frequently compute adjoints to study the estimate of sensitivity of model output with respect to input. Factors to be considered in adjoint computation include the accuracy of derivatives, the speed of computing them, and the ease with which they can be computed. This minisymposium will present talks showing examples of adjoint computation in earth system models as well as advances in their efficient computation.
Organizer: Sri Hari Krishn Narayanan Argonne National Laboratory, USA
Organizer: Paul D. Hovland Argonne National Laboratory, USA 10:00-10:20 Improving the Efficiency of the Adjoint of Fixed-Point Iterations Sri Hari Krishn Narayanan, Argonne National Laboratory, USA; Daniel Goldberg, University of Edinburgh, United Kingdom; Paul D. Hovland, Argonne National Laboratory, USA
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
Tuesday, March 17
MS191
MS192
Numerical Methods for High-Dimensional Stochastic and Parametric Problems - Part I of V 10:00 AM-11:40 AM
The System Dynamics of Social and Health Processes using Quantitative Data Sciences Methods 10:00 AM-11:40 AM
Room:251 B
Room:251 C
For Part 2 see MS215 Development of scalable numerical methods for the solution of problems with highdimensional stochastic or parametric inputs has been a subject of active research in computational sciences and engineering. This is motivated by the need to reduce the issue of curse-of-dimensionality, i.e., exponential increase of computational complexity, in predictive simulation of physical systems where accurate specification of governing laws entails a large number of parameters or stochastic variables. To this end, several novel approaches based on multi-level, reduced order, sparse, and lowrank approximations have been recently developed. This minisymposium presents state-of-the-art in such developments for various aspects of high-dimensional computation, including analysis, algorithms, implementation, and applications.
Part of the SIAM Workshop Celebrating Diversity
Organizer: Alireza Doostan University of Colorado Boulder, USA
Organizer: Dongbin Xiu
Recent years have witnessed an increasing interest in developing models for understanding imminent challenges in engineering and public health. In this minisymposium, the talks will highlight some of the issues encountered when trying to apply data-driven approaches to the real-time systems. The increasing availability of data from a variety of sources including social media provides the ability to model complex ecosystems enabling human decision-making. Models have the potential to facilitate more accurate assessment for such systems, and to provide a basis for more efficient and targeted approaches to treatment and scheduling, through an improved understanding of the mechanisms of action.
Organizer: Anuj Mubayi Arizona State University, USA
Organizer: Raegan Higgins Texas Tech University, USA
Organizer: Susan E. Minkoff
10:25-10:45 Development of a TimeDependent Ice Flow Model Adjoint and Its Applications Daniel Goldberg, University of Edinburgh, United Kingdom; Patrick Heimbach, Massachusetts Institute of Technology, USA
University of Utah, USA 10:00-10:20 Local Polynomial Chaos Methods for High Dimensional SPDE Yi Chen, Purdue University, USA; John D. Jakeman, Sandia National Laboratories, USA; Xueyu Zhu and Dongbin Xiu, University of Utah, USA
University of Texas at Dallas, USA
10:50-11:10 Parallel 4D Variational Data Assimilation Vishwas Hebbur Venkata Subba Rao and Adrian Sandu, Virginia Tech, USA
10:25-10:45 Uncertainty Propagation Using Infinite Mixture of Gaussian Processes and Variational Bayesian Inference Peng Chen and Nicholas Zabaras, Cornell University, USA; Ilias Bilionis, Purdue University, USA
10:25-10:45 The Collective Impact of Social Factors and Interventions on the Dynamics of Reported NarcoticRelated Criminal Cases in the Community Areas of Chicago Maryam Khan, Arizona State University, USA
11:15-11:35 An Adjoint Based Analysis of the Physical Drivers of Uncertainty in Air-Sea Exchange and Ocean Draw Down of Co2 Chris Hill and Oliver Jahn, Massachusetts Institute of Technology, USA; Jean Utke, Argonne National Laboratory, USA
10:50-11:10 Title Not Available at Time of Publication Raul F. Tempone, King Abdullah University of Science & Technology (KAUST), Saudi Arabia 11:15-11:35 Multivariate Weighted Least-squares using Monte Carlo Samples Akil Narayan, University of Massachusetts, Dartmouth, USA
Organizer: Stephen Wirkus Arizona State University, USA 10:00-10:20 Analysis of Information Diffusion on Social Networks Daniel Romero, University of Michigan, USA
10:50-11:10 An Effective Communitybased Approach to Mitigate Sybil Attacks in Online Social Networks Satyajayant Misra, New Mexico State University, USA 11:15-11:35 The Dynamics of Co-Evolution of Health Behaviors in College Population Anuj Mubayi, Arizona State University, USA
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS193 Computational Benchmarks for Soft Materials and Complex Fluids at the Mesoscale - Part I of III 10:00 AM-11:40 AM Room:251 D For Part 2 see MS217 Computational modeling of multiphysical processes at the mesoscale is becoming increasingly important for applications in micro-/nano-technology, next-generation battery technology, material assembly and biological systems. In this minisymposium we address mathematical modeling along with numerical schemes that can effectively capture mesoscale multiphysics such as hydrodynamics, transport, electrostatics and chemical reaction. Desirable features of the numerical schemes include robustness, accuracy, scalability, as well as potential to cross scales as a multiscaling framework. The benchmark applications that we are interested in include mixing and separation processes in micro-/nano-channel, semipermeable membranes, diffusive transport and electrochemical reaction in porous electrode, colloid and nanoparticle assembly, as well as macromolecular conformational change.
10:50-11:10 Fluctuating Hydrodynamics of Suspensions of Rigid Particles Aleksandar Donev, Courant Institute of Mathematical Sciences, New York University, USA 11:15-11:35 SPH Model for LandauLifshitz Navier-Stokes and AdvectionDiffusion Equations Alexander Tartakovsky, Pacific Northwest National Laboratory, USA; Jannes Kordilla, University of Goettingen, Germany; Wenxiao Pan, Pacific Northwest National Laboratory, USA
125
Tuesday, March 17
MS194 Inverse Problems and Data Assimilation - Part II of VI 10:00 AM-11:40 AM Room:251 E For Part 1 see MS168 For Part 3 see MS218 The aim of this minisymposium is to document recent mathematical developments in the field of inverse problems and data assimilation based on theoretical and numerical grounds that are relevant for various scientific and real life applications. Topics for the minisymposium include but are not limited to: 1) deterministic and statistical approaches to inversion problems and data assimilation; 2) reduced order modeling inversion framework; 3) treatment of nonlinear and non-smooth processes; 4) sensitivity analysis and its application to adaptive (targeting) observation; 5) operational data assimilation systems; 6) uncertainties impact studies; 7) automatic tools to support inversion and data assimilation methodologies.
Organizer: Razvan Stefanescu
Organizer: Mauro Perego
Virginia Tech, USA
Sandia National Laboratories, USA
Organizer: Adrian Sandu
Organizer: Wenxiao Pan
Virginia Tech, USA
Pacific Northwest National Laboratory, USA
Organizer: Ionel M. Navon
Organizer: Nathaniel Trask
Florida State University, USA
Brown University, USA
Organizer: Dacian N. Daescu
Organizer: Kyungjoo Kim
Portland State University, USA
Sandia National Laboratories, USA
10:00-10:20 Toward New Applications of the Adjoint Tools in 4D-Var Data Assimilation Dacian N. Daescu, Portland State University, USA; Ricardo Todling, NASA, USA; Rolf Langland, Naval Research Laboratory, USA; Austin Hudson, Portland State University, USA
10:00-10:20 A Nano Pore-Scale Model for the Nanostructured Cathode of Lithium-Oxygen Batteries Wenxiao Pan, Pacific Northwest National Laboratory, USA 10:25-10:45 Boltzmann’s State of Motion: Phenomenological Modeling of Chemical and Ecological Systems Yian Ma and Hong Qian, University of Washington, USA
continued in next column
10:25-10:45 Computing Exactly and Eficiently Arbitrarily-High-Order Response Sensitivities to Model Parameters Dan G. Cacuci, University of South Carolina, USA
continued on next page
126
Tuesday, March 17
MS194 Inverse Problems and Data Assimilation - Part II of VI 10:00 AM-11:40 AM continued
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
Tuesday, March 17
MS195
MS196
Uncertainty Quantification Methods for Power Grid Systems - Part I of II 10:00 AM-11:40 AM
Numerical Methods for Multimaterial Flows with Deforming Boundaries Part I of III 10:00 AM-11:40 AM
Room:251 F 10:50-11:10 Dealing with Nonsmoothness in Data Assimilation Andreas Griewank, Humboldt University Berlin, Germany 11:15-11:35 Second Order Analysis in Variational Data Assimilation Francois-Xavier L. Le-Dimet, Université de Grenoble Alpes, France; M Yousuff Hussaini, Florida State University, USA; Ha Tran Thu, Vietnamese Academy of Science and Technology, Vietnam
For Part 2 see MS219 The properties of many operational and structural elements of power grid models are typically known to some degree of certainty. Examples include grid topology and state, including knowledge of what lines are down; model parameters, such as generator/load/line electrical and performance characteristics; and operational conditions, such as renewable generator and load levels. In this minisymposium we bring together modeling contributions for the treatment of these uncertainties, to improve the design and operation of power grids.
Room:254 A
Sandia National Laboratories, USA
For Part 2 see MS220 Multimaterial flows with deforming boundaries appear in many applications involving biological flows, medicine, industry, and homeland security. As the number of applications grow, so too the amount of research into developing and improving existing numerical methods for deforming boundary problems in multi-material flows. In this minisymposium, novel numerical methods will be presented that address efficiency and accuracy in adaptive mesh refinement algorithms, scalability on high performance computing platforms, robust interface capturing methods, numerical methods for phase change (freezing, boiling, evaporation), and numerical methods for enforcing boundary conditions at deforming boundary interfaces.
Organizer: Jean-Paul Watson
Organizer: Mark Sussman
Organizer: Cosmin Safta Sandia National Laboratories, USA
Organizer: Ali Pinar
Sandia National Laboratories, USA 10:00-10:20 An Efficient Approach for Stochastic Optimization of Electricity Grid Operations Cosmin Safta, Habib N. Najm, Richard L. Chen, Ali Pinar, and Jean-Paul Watson, Sandia National Laboratories, USA 10:25-10:45 Two-Stage Adaptive Robust Unit Commitment Using Scenarios Induced Uncertainty Set Richard L. Chen, Cosmin Safta, Jean-Paul Watson, Habib N. Najm, and Ali Pinar, Sandia National Laboratories, USA 10:50-11:10 Data-Driven Model for Solar Irradiation Based on Satellite Observations Ilias Bilionis, Purdue University, USA; Emil M. Constantinescu and Mihai Anitescu, Argonne National Laboratory, USA 11:15-11:35 Economic Impacts of Wind Covariance Estimation on Power Grid Operations Cosmin G. Petra and Victor Zavala, Argonne National Laboratory, USA; Elias NinoRuiz, Virginia Tech, USA; Mihai Anitescu, Argonne National Laboratory, USA
Florida State University, USA
Organizer: Frederic G. Gibou University of California, Santa Barbara, USA
Organizer: Feng Xiao Tokyo Institute of Technology, Japan 10:00-10:20 High Resolution PDE Solvers on Octree Grids and Parallel Architectures Frederic G. Gibou, University of California, Santa Barbara, USA 10:25-10:45 Topological Change with a Cut Cell based Sharp Interface Method for Multi-phase Flows Xiao-Long Deng, Beijing Computational Science Research Center, China 10:50-11:10 A Second Order Virtual Node Algorithm for Navier–Stokes Flow Problems with Interfacial Forces and Discontinuous Material Properties Joseph Teran, University of California, Los Angeles, USA 11:15-11:35 On the Coupling of FarField Wind-Wave Simulation and NearField Free-Surface Flow Simulation Antoni E. Calderer, University of Minnesota, USA; Xin Guo, University of California, Berkeley, USA; Fotis Sotiropoulos and Lian Shen, University of Minnesota, USA
2015 SIAM Conference on Computational Science and Engineering
127
Tuesday, March 17
Tuesday, March 17
Tuesday, March 17
MS197
MS198
MS199
Parallel Computing for Partial Differential Equations on CPUs, GPUs, and Intel Phi - Part I of II 10:00 AM-11:40 AM
Student Days: Combining Disciplines, Techniques, Faculty and Students to Tackle Protein Folding 10:00 AM-11:40 AM
Hybrid and Multilevel Approaches to Kinetic Equations - Part III of IV 10:00 AM-11:40 AM
Room:254 B
Room:254 C
For Part 2 see MS221 Modern architectures including multi-core CPUs, massively parallel GPUs, and manycore accelerators such as the 60-core Intel Phi processor offer great opportunities for speeding up computations. They pose significant challenges due to the hybrid programming models and libraries that need to be used, ranging from distributed-memory computing with MPI, multi-threading with OpenMP, CUDA on GPUs, offloading to the Phi, for instance. This minisymposium will share experiences with parallel code and numerical algorithms from fundamental test problems over established benchmark codes to specialized research code. The minisymposium will start with an overview and contrast of the architectures and include the opportunity for the audience to discuss their own experiences.
Consistently predicting protein structure is a grand challenge in biology. This symposium features speakers from an educational experiment that was conducted during the summer of 2014: nine students and three faculty representing different institutions and disciplines, spent 10 weeks under one roof discussing protein folding and learning about HPC. The WeFold project that brings together researchers worldwide to collaborate and compete at the CASP protein folding competition, served as the educational framework. The learning curve for the students was steep and the results impressive suggesting that this approach is an effective methodology to introduce students to computational sciences and HPC.
For Part 2 see MS173 For Part 4 see MS223 In recent years, a significant amount of effort has been directed to solving kinetic equations. Generally, kinetic equations have posed a challenge due to the large phase space associated with the equations, however algorithmic advances and advances in computer hardware have made these simulations more tractable. In this minisymposium, we consider hybrid and multilevel approaches to solving kinetic equations.
Organizer: Matthias K. Gobbert University of Maryland, Baltimore County, USA
Organizer: Ulrich J. Ruede University of Erlangen-Nuremberg, Germany 10:00-10:20 Overview and Contrast of Modern Computer Architectures Including the Intel Phi Jonathan Graf, Samuel Khuvis, Xuan Huang, and Matthias K. Gobbert, University of Maryland, Baltimore County, USA 10:25-10:45 The HPCG Benchmark Using Intel Phi Accelerators Adam Cunningham, Gerald Payton, and Jack Slettebak, University of Maryland, Baltimore County, USA; Jordi WolfsonPou, University of California, Santa Cruz, USA 10:50-11:10 Offloading Computational Kernels in Long-Time Simulations to the Intel Phi Samuel Khuvis, Xuan Huang, Jonathan Graf, and Matthias K. Gobbert, University of Maryland, Baltimore County, USA 11:15-11:35 Asynchronous Preconditioning on Accelerators Edmond Chow, Georgia Institute of Technology, USA
Room:150 AB
Organizer: Jeffrey A. Willert Los Alamos National Laboratory, USA
Organizer: Cory Hauck Oak Ridge National Laboratory, USA
Organizer: Silvia N. Crivelli
Organizer: Ryan McClarren
Lawrence Berkeley National Laboratory, USA
Texas A&M University, USA
Organizer: Martin Frank
10:00-10:20 WeFold: a Collaborative and Educational Experiment Silvia N. Crivelli, Lawrence Berkeley National Laboratory, USA; John Hatherill, Del Mar Community College, USA; Jesse Fox, Mountain View Community College, USA
RWTH - Aachen University of Technology, Germany
10:25-10:45 Creating the Largest Decoys Database to Improve Scoring Functions Using Machine Learning Ricardo Ferreira and Christopher Cook, Mountain View Community College, USA 10:50-11:10 The Maintenance of the WeFold Gateway for CASP11 Anthony Lopez, Del Mar Community College, USA 11:15-11:35 Reducing the Data Complexity with Filtering and Clustering Rachel A. Davis, Drake University, USA; Jennifer Ogden, Saint Mary’s College of California, USA; Rehan Raiyyani, University of California, San Diego, USA
Organizer: Jingmei Qiu University of Houston, USA 10:00-10:20 A Moment Model for the Vlasov Fokker Planck Equation Charles K. Garrett and Cory Hauck, Oak Ridge National Laboratory, USA 10:25-10:45 An Implcit, Conservative Vlasov-Darwin Pic Solver in Multiple Dimensions Guangye Chen and Luis Chacon, Los Alamos National Laboratory, USA 10:50-11:10 Modeling Non-Ideal Plasmas: a Hyrbid Quantum Hydrodynamics and Molecular Dynamics Approach Michael Murillo, Los Alamos National Laboratory, USA 11:15-11:35 iFP: An Optimal, Fully Conservative, Fully Implicit, VlasovFokker-Planck Solver William T. Taitano, Luis Chacon, and Andrei Simakov, Los Alamos National Laboratory, USA
128
Tuesday, March 17
MS200 Partitioning and Load Balancing in Scientific Applications 10:00 AM-11:40 AM Room:150 DE The importance of partitioning and load balancing is growing with the emergence of computers with 100K+ cores. This minisymposium will address issues involved with partitioning and load balancing at extreme scale for scientific applications. These issues include partitioning for extreme core counts, for memory-constrained processors, and for complex objectives.
Organizer: Karen D. Devine Sandia National Laboratories, USA
Organizer: Siva Rajamanickam Sandia National Laboratories, USA
Organizer: Brian Van Straalen Lawrence Berkeley National Laboratory, USA 10:00-10:20 Unstructured Mesh Partitioning to over 500k Parts Cameron Smith, Dan A. Ibanez, and Mark S. Shephard, Rensselaer Polytechnic Institute, USA 10:25-10:45 Current Challenges in Mesh Partitioning for Physics Simulations Cédric Chevalier, CEA, France 10:50-11:10 The Zoltan2 Toolkit: Partitioning, Task Placement, Coloring, and Ordering Karen D. Devine, Erik G. Boman, Siva Rajamanickam, and Lee Ann Riesen, Sandia National Laboratories, USA; Mehmet Deveci and Umit V. Catalyurek, The Ohio State University, USA 11:15-11:35 Zoltan2 for Extreme-Scale Data Partitioning Michael Wolf, Sandia National Laboratories, USA
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS201 Deployment and Application of Technologies provided by the FASTMath Institute - Part I of II 10:00 AM-11:40 AM Room:258 For Part 2 see MS225 The FASTMath SciDAC Institute develops and deploys scalable mathematical algorithms and software tools for reliable simulation of complex physical phenomena. These tools include structured and unstructured mesh techniques, linear and nonlinear solvers, eigensolvers and time integration methods. Research efforts are focused on developing a full range of accurate and robust technologies that support different application codes specifically those employed by the Department of Energy’s domain scientists. This minisymposium will place emphasis on application of FASTMath tools on SciDAC and related applications. Speakers will discuss the challenges in different application areas and the impact of the latest developments in FASTMath tools.
Organizer: Onkar Sahni Rensselaer Polytechnic Institute, USA
Organizer: Lori A. Diachin Lawrence Livermore National Laboratory, USA 10:00-10:20 Gyrokinetic Poisson Equation Solvers with Explicit Flux Surface Averaging in XGC1 with PETSc Mark Adams, Lawrence Berkeley National Laboratory, USA; Seung-Hoe Ku, Princeton Plasma Physics Laboratory, USA 10:25-10:45 Preconditioners for Implicit Atmospheric Climate Simulations in the Community Atmosphere Model David J. Gardner, Lawrence Livermore National Laboratory, USA; Katherine J. Evans, Oak Ridge National Laboratory, USA; Aaron Lott, D-Wave Systems, Inc., Canada; Andrew Salinger, Sandia National Laboratories, USA; Carol S. Woodward, Lawrence Livermore National Laboratory, USA
continued in next column
10:50-11:10 Algebraic Multigrid Solvers for Lattice QCD in the Hypre Software Library Evan Berkowitz, Lawrence Livermore National Laboratory, USA; James Brannick, Pennsylvania State University, USA; Robert Falgout and Chris Schroeder, Lawrence Livermore National Laboratory, USA 11:15-11:35 Block-Structured AMR: Applications Using BoxLib Ann S. Almgren, Lawrence Berkeley National Laboratory, USA
2015 SIAM Conference on Computational Science and Engineering
129
Tuesday, March 17
Tuesday, March 17
Tuesday, March 17
MS202
MS203
MS204
Advances in Multigrid Methods and Their Applications - Part I of II 10:00 AM-11:40 AM
Software Components for Integral Equation Methods Part III of IV 10:00 AM-11:40 AM
Room:259
Room:260 A
Recent Advances in High Order Spatial Discretization Methods for PDEs Part I of III 10:00 AM-11:40 AM
For Part 2 see MS226 Multigrid is the most successful and generally-applicable scalable approach to solving problems with a hierarchy of scales in application areas such as elliptic PDEs, optimization, inversion, and image processing. Scalable methods are increasingly essential as hardware evolution and algorithmic advances enable the solution of larger scale problems in science and engineering applications. This session brings together researchers developing multigrid and related methods to assess the state of the art, recent theoretical advances, and novel applications.
Organizer: Maria Emelianenko
For Part 2 see MS177 For Part 4 see MS227 Due to recent advances in the underlying technology, integral equations have become highly competitive in attacking the most challenging of elliptic, globally-coupled PDE problems. However, compared to a standard off-the-shelf finite element methods, a larger amount of machinery is involved, including fast algorithms, preconditioners, and singular quadrature methods. To make this theoretically attractive family of methods practically usable, many of these components need to be readily available in software form. This minisymposium seeks to bring together practitioners and researchers with the goal of encouraging exchange of ideas and discovering opportunities for collaboration in this direction.
George Mason University, USA
Organizer: Timo Betcke
Organizer: Zichao Di Argonne National Laboratory, USA
Organizer: Jed Brown
University College London, United Kingdom
Argonne National Laboratory, USA and University of Colorado Boulder, USA
Organizer: Andreas Kloeckner
10:00-10:20 Optimization Approach for Tomographic Inversion from Multiple Data Modalities Zichao Di, Sven Leyffer, and Stefan Wild, Argonne National Laboratory, USA
10:00-10:20 Rank-Structured Preconditioners for Two and ThreeDimensional Integral and Differential Equations Steffen Börm and Knut Reimer, Kiel University, Germany
10:25-10:45 Multigrid Preconditioning for Space-time Distributed Optimal Control Problems Constrained by Parabolic Equations Andrei Draganescu and Mona Hajghassem, University of Maryland, Baltimore County, USA 10:50-11:10 Multigrid in Chaos Qiqi Wang and Patrick Blonigan, Massachusetts Institute of Technology, USA 11:15-11:35 About Some Smoothers for Saddle-point Problems Carmen Rodrigo, Francisco José Gaspar, and Francisco Lisbona, University of Zaragoza, Spain; Ludmil Zikatanov, Pennsylvania State University, USA
University of Illinois, USA
10:25-10:45 Practical and Efficient Direct Solvers for BIEs Gunnar Martinsson, University of Colorado Boulder, USA 10:50-11:10 Fast Solvers for Hierarchical Matrices Amirhossein Aminfar, Stanford University, USA; Sivaram Ambikasaran, Courant Institute of Mathematical Sciences, New York University, USA; Mohammad Hadi Pour Ansari and Eric F. Darve, Stanford University, USA 11:15-11:35 A New Integral Formulation and Fast Direct Solver for Periodic Stokes’ Flow Adrianna Gillman, Rice University, USA; Alex H. Barnett, Dartmouth College and Simons Foundation, USA; Shravan Veerapaneni and Gary Marple, University of Michigan, USA
Room:260 B For Part 2 see MS228 Many complex physical phenomena can be modeled by partial differential equations and the numerical simulation of PDEs has been an active research area. This minisymposium aims to bring researchers together to discuss the numerical challenges when designing high order spatial discretization methods for time-dependent/static PDEs, with particular attention on weighted essentially non-oscillatory (WENO) methods and discontinuous Galerkin (DG) methods.
Organizer: Yuan Liu Michigan State University, USA
Organizer: Qi Tang Michigan State University, USA
Organizer: Andrew J. Christlieb Michigan State University, USA 10:00-10:20 High Order WENO Method for Steady State Problems Liang Wu and Yongtao Zhang, University of Notre Dame, USA; Shuhai Zhang, China Aerodynamics Research and Development Center, China; Chi-Wang Shu, Brown University, USA 10:25-10:45 A Hybrid Weno Reconstruction on Unstructured Mesh Yuan Liu, Michigan State University, USA 10:50-11:10 A New RKDG Method with Conservation Constraint to Improve CFL Condition for Solving Conservation Laws Zhiliang Xu, University of Notre Dame, USA 11:15-11:35 A High-Order Adaptive Finite Volume Solver for Steady Euler Equations Guanghui Hu, University of Macau, China
130
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
Tuesday, March 17
MS206
MS207
Advances in Computational Techniques for Coastal Ocean Modeling Part I of II 10:00 AM-11:40 AM
Advances in High Level Finite Element Systems Part I of II 10:00 AM-11:40 AM
The theme of this minisymposium is the optimization of complex engineering systems that are modelled using computationallyexpensive PDE simulations. The focus is on gradient-based algorithms applied to reduced-space formulations, in which the state variables are considered implicit functions of the design via the PDE constraint. Compared with full-space, or "one-shot,” methods, reduced-space formulations offer a simplified and modular approach to optimization. However, despite its long history and popularity, the reducedspace approach continues to present unique challenges, which the speakers will discuss and address.
Room:151 DE For Part 2 see MS230 Environmental and coastal ocean engineering models yield complex systems combining interdisciplinary techniques. Moreover, their accurate and efficient simulation requires advanced techniques in high performance scientific computing. This minisymposium brings together experts to discuss a number of new ideas for the development, analysis and application of this important class of problems. Specific ideas include the high resolution modeling of weather extremes such as hurricane storm surge, wave propogation, climate change, and flood control.
For Part 2 see MS231 Many modern software projects utilize the high level of mathematical abstraction provided by finite element methods to develop powerful libraries that hide or even automate many practical computational details. Such codes increase programmer productivity, decrease potential sources of errors, and simplify the implementation of complex calculations such as h/p adaptivity and adjoint-based optimization. This minisymposium will bring together key participants from Deal.II, Dune, FEniCS, Dolfin-adjoint, Firedrake and Nektar++ to communicate the latest advances in and applications of such finite element systems.
Organizer: Jason E. Hicken
Organizer: Jennifer Proft
Organizer: David Ham
Tuesday, March 17
MS205 Reduced-space PDEconstrained Optimization for Engineering Design 10:00 AM-11:40 AM Room:151 AB
Rensselaer Polytechnic Institute, USA
Organizer: Graeme Kennedy Georgia Institute of Technology, USA 10:00-10:20 Constraint Aggregation Methods for PDE-Constrained Optimization Graeme Kennedy, Georgia Institute of Technology, USA; Jason E. Hicken, Rensselaer Polytechnic Institute, USA 10:25-10:45 A Krylov-Based Iterative Solver for Equality-Constrained NonConvex Quadratic Subproblems Jason E. Hicken and Alp Dener, Rensselaer Polytechnic Institute, USA 10:50-11:10 Aerodynamic Shape Optimization with Goal-Oriented Error Estimation and Control Marian Nemec and Michael Aftosmis, NASA Ames Research Center, USA 11:15-11:35 Large-Scale PDEConstrained Fluid-Structure Optimization Gaetan Kenway and Joaquim R. R. A Martins, University of Michigan, USA
University of Texas at Austin, USA
Organizer: Clint Dawson University of Texas at Austin, USA 10:00-10:20 A Parallel Local Timestepping Runge-Kutta Discontinuous Galerkin Method with Applications to Coastal Ocean Modeling Clint Dawson, University of Texas at Austin, USA 10:25-10:45 Physically Based Assessment of Hurricane Surge Threat under Climate Change Ning Lin, Princeton University, USA 10:50-11:10 Understanding Coastal Hydrodynamic Processes and Mitigating Risk Through High Fidelity Computer Simulations Joannes Westerink, University of Notre Dame, USA 11:15-11:35 Development and Validation of DG Wave: a Discontinuous Galerkin-Based Numerical Wave Prediction Model Ethan Kubatko, The Ohio State University, USA
Room:151 G
Imperial College London, United Kingdom
Organizer: Rob Kirby Baylor University, USA 10:00-10:20 FinAT: A Mathematical Structure-Preserving Library of Finite Elements David Ham, Imperial College London, United Kingdom; Rob Kirby, Baylor University, USA 10:25-10:45 Multicore Parallelism for Common Finite Element Operations Bruno Turcksin, Texas A&M University, USA; Martin Kronbichler, Technische Universität München, Germany; Wolfgang Bangerth, Texas A&M University, USA 10:50-11:10 Towards a Unified Framework for Automated a Posteriori Error Estimation and Adaptivity in Space-Time Marie E. Rognes, Simula Research Laboratory, Norway; Anders Logg, Chalmers University of Technology, Sweden; Benjamin Kehlet, Simula Research Laboratory, Norway 11:15-11:35 Finite Element Geometric Multigrid Solvers from High-Level Problem Descriptions Lawrence Mitchell, Imperial College London, United Kingdom; Eike H. Mueller, University of Bath, United Kingdom; David Ham and Colin J. Cotter, Imperial College London, United Kingdom
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17 Intermission 11:40 AM-11:50 AM
IP8 The Power of Matrix and Tensor Decompositions in Smart Patient Monitoring 11:50 AM-12:35 PM Room:355 Chair: Tom Bartol, The Salk Institute, USA Accurate and automated extraction of clinically relevant information from patient recordings requires an ingenious combination of adequate pretreatment of the data (e.g. artefact removal), feature selection, pattern recognition, decision support, up to their embedding into user-friendly user interfaces. The underlying computational problems can be solved by making use of matrix and tensor decompositions as building blocks of higher-level signal processing algorithms. A major challenge here is how to make the mathematical decompositions 'interpretable’ such that they reveal the underlying medically relevant information and improve medical diagnosis. The application of these decompositions and their benefits will be illustrated in a variety of case studies, including epileptic seizure onset localisation using adult and neonatal scalp EEG and Event-related potential analysis during simultaneous EEG-fMRI acquisition.
Sabine Van Huffel Katholieke Universiteit Leuven, Belgium
Lunch Break 12:35 PM-2:15 PM Attendees on their own
Workshop Celebrating Diversity (WCD) Luncheon (by invitation only) 12:45 PM-2:15 PM Room:255 D
131
Tuesday, March 17
Tuesday, March 17
PD4 Panel
PD5
Data Science: What is It and How to Teach It 12:45 PM-2:15 PM
Thinking of Writing a Book? 12:45 PM-2:15 PM
Room:355 A
Chair: Nicholas J. Higham, University of Manchester, United Kingdom
Chair: Christopher Johnson, University of Utah, USA Chair: Jeff Phillips, University of Utah, USA Recently, the term Big Data has become ubiquitous. People who can wrangle Big Data are called Data Scientists. According to a number of sources, there is a growing need for people trained as Data Scientists. But what is Data Science? Is Data Science its own field or is it an interdisciplinary mix of Computer Science, Mathematics and Statistics, and Domain Knowledge or perhaps it is really what Statisticians have been doing all along? Since data science at scale involves large-scale computation, what is the relation between data science and computational science in research and education? This panel features leaders in Data Science and Computational Science to discuss the current and future status of Data Science, its relationship to Computational Science, opportunities for Data and Computational Scientists and educating future Data Scientists.
Panelists: To Be Announced
Room:355 D
Ever thought about writing a book? Ever wondered just what that entails? This session brings together successful authors and publishing staff to discuss the process. Topics of interest will include: •
Why and when you should consider writing a book • A step-by-step description of the process, from initial idea to published book • How to choose a publisher • The author/publisher relationship – who does what • Pitfalls to avoid • Tips from successful authors Even if you aren’t currently thinking of becoming an author, this session promises to be lively and engaging! Daniela Calvetti Case Western Reserve University, USA Eldad Haber Emory University, USA Ralph Smith North Carolina State University, USA Elizabeth Greenspan SIAM, USA David Marshall SIAM, USA
132
Tuesday, March 17
MT3 Minitutorial: Lab Skills for Scientific Computing Part I of II 2:15 PM-3:55 PM Room:355 For Part 2 see MT4 Chair: Greg Wilson, Mozilla Foundation, USA The Software Carpentry project (http:// software-carpentry.org) has been teaching basic computing skills to scientists and engineers since 1998. This minitutorial will introduce the tools and techniques that have proven most useful, and show how integrating them can help researchers get more done in less time, and with less pain. This two-part minitutorial will introduce several widely-used practices in software development, explore the empirical evidence showing their benefits (or in some cases the lack thereof), and describe how researchers and research teams can adopt them. Some of the work will be hands-on, so participants are strongly urged to bring a laptop. Warning: real-world examples may be used. Greg Wilson, Mozilla Foundation, USA
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
Tuesday, March 17
MS208
MS209
Recent Advances in Numerical Methods for Interface Problems Part I of IV 2:15 PM-3:55 PM
First Principle Methods and Applications for Computational Materials Science and ChemistryPart II of III 2:15 PM-3:55 PM
Room:250 A For Part 2 see MS233 Many real world applications are characterized by multiple materials and complex fluids, and they often lead to interface problems. It is imperative to develop efficient and stable numerical methods for these problems. Great efforts have been made for solving interface problems and tracing the moving interfaces in the past decades. However, many challenges, such as the lack of regularity of physical solutions due to the discontinuity across interfaces, still hinder the development of efficient numerical methods. This minisymposium intends to create a forum for researchers from different fields to discuss recent advances on the interface methods and their applications.
Organizer: Xiaoming He Missouri University of Science and Technology, USA
Organizer: Xiaolin Li State University of New York, Stony Brook, USA 2:15-2:35 Modelling of Fabric Surface for Parachute Inflation through Front Tracking Xiaolin Li, State University of New York, Stony Brook, USA 2:40-3:00 Thermodynamically Consistent and Meta-Stable Equation of State Models for Hydro and Solid Dynamics John W. Grove, Los Alamos National Laboratory, USA 3:05-3:25 Overlapping BEM on FEM computations Francisco J. J. Sayas, University of Delaware, USA; Victor Dominguez, Universidad Pública de Navarra, Spain; Matthew Hassell, University of Delaware, USA 3:30-3:50 An Augmented Method for Stokes-Darcy Coupling and Applications Zhilin Li, North Carolina State University, USA
Room:250 B For Part 1 see MS184 For Part 3 see MS234 First principle electronic structure calculations are among the most challenging and computationally demanding science and engineering problems. This minisymposium aims at presenting and discussing new numerical methods for achieving ever higher level of accuracy and efficiency in ground and excited density functional theory calculations, and their applications in computational materials science and chemistry. We propose to bring together physicists, chemists and materials scientists who are involved in improving the numerical development of widely known quantum chemistry and solid-state physics application software packages, with mathematicians and computer scientists who are focusing on advancing the required state-of-theart mathematical algorithms and parallel implementation.
Organizer: Lin Lin University of California, Berkeley, USA
Organizer: Eric Polizzi University of Massachusetts, Amherst, USA
Organizer: Chao Yang Lawrence Berkeley National Laboratory, USA 2:15-2:35 Truly Scalable O(N) Approach for First-Principles Molecular Dynamics (FPMD) of NonMetallic Systems Jean-Luc Fattebert and Daniel Osei-Kuffuor, Lawrence Livermore National Laboratory, USA 2:40-3:00 Towards Ab-Initio Simulations of Nanoelectronic Devices Mathieu Luisier, Mauro Calderara, Sascha Brueck, Hossein Bani-Hashemian, and Joost VandeVondele, ETH Zürich, Switzerland
continued on next page
2015 SIAM Conference on Computational Science and Engineering
3:05-3:25 Using Next-generation Architectures to Model Large and Complex Molecular Environments Bert de Jong, Hongzhang Shan, and Leonid Oliker, Lawrence Berkeley National Laboratory, USA 3:30-3:50 Ab Initio Quantum Monte Carlo in Computational Materials Science and Chemistry Paul Kent, Oak Ridge National Laboratory, USA
133
Tuesday, March 17
Tuesday, March 17
MS210
MS211
Resilient, Extreme-scale Numerical Algorithms Part II of II 2:15 PM-3:55 PM
Recent Advances in Model Reduction - Part II of V 2:15 PM-3:55 PM
Room:250 C
For Part 1 see MS187 For Part 3 see MS236 Model reduction has become an increasingly important tool to mitigate the computational burden of modeling and simulation in timecritical (e.g., model-predictive control) and many query (e.g., Bayesian inversion) applications. This minisymposium presents recent advances that address the primary challenges facing such methods, such as preserving intrinsic problem structure, handling high-dimensional parameter spaces, integration with existing simulation codes, stability, optimization and uncertainty quantification and control.
For Part 1 see MS186 Computer architectures are undergoing fundamental changes that will require the rethinking and invention of numerical algorithms suitable for computing at extreme scales (over 100 PFlops). Numerical solver technologies are needed that favor computation over data motion, that expose much greater concurrency, and that are resilient to increasing soft and hard faults. This minisymposium will feature progress in the development of such algorithms by several projects funded by the DOE Resilient Extreme-Scale Solvers initiative. A breadth of topics will be addressed, including algebraic multigrid, methods for hyperbolic conservation laws, Particle-In-Cell algorithms, and stochastic methods for linear systems.
Organizer: John Loffeld Lawrence Livermore National Laboratory, USA
Organizer: Andrew Myers Lawrence Berkeley National Laboratory, USA
Organizer: Jeffrey A. Hittinger Lawrence Livermore National Laboratory, USA 2:15-2:35 Parallel Algorithms for the Monte Carlo Synthetic Acceleration Linear Solver Method Stuart Slattery, Tom Evans, and Steven Hamilton, Oak Ridge National Laboratory, USA 2:40-3:00 Iterative Performance of Monte Carlo Linear Solver Methods Massimiliano Lupo Pasini, Emory University, USA 3:05-3:25 Comparative Performance Analysis of an Algebraic Multigrid Solver on Leading Multicore Architectures Alex Druinsky, Brian Austin, Xiaoye Sherry Li, Osni A. Marques, Eric Roman, and Samuel Williams, Lawrence Berkeley National Laboratory, USA 3:30-3:50 Segmental Refinement: A Multigrid Technique for Data Locality Mark Adams, Lawrence Berkeley National Laboratory, USA
Room:250 D
Organizer: Kevin T. Carlberg Sandia National Laboratories, USA
Organizer: Gianluigi Rozza SISSA, International School for Advanced Studies, Trieste, Italy 2:15-2:35 Reduced Order Modeling of Geophysical Flows Traian Iliescu, Virginia Tech, USA 2:40-3:00 Energy-based Inner Products for POD/Galerkin Model Reduction for Compressible Flows Jeffrey Fike, Irina Kalashnikova, Matthew Barone, and Srinivasan Arunajatesan, Sandia National Laboratories, USA 3:05-3:25 Data-driven Optimal Rational Approximation via Numerical Quadrature Christopher A. Beattie, Virginia Tech, USA; Zlatko Drmac, University of Zagreb, Croatia; Serkan Gugercin, Virginia Tech, USA 3:30-3:50 Efficient Reduced Basis Methods for Contact and Related Problems Karen Veroy-Grepl, Zhenying Zhang, Eduard Bader, and Mark Kaercher, RWTH Aachen University, Germany
134
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
Tuesday, March 17
Tuesday, March 17
MS212
MS213
MS214
Advances in Time-stepping Methods - Part IV of V 2:15 PM-3:55 PM
Sparse and Compressible Representations: Theory, Algorithms and Applications - Part II of III 2:15 PM-3:55 PM
Efficient Algorithms for Variational Methods in Imaging - Part I of II 2:15 PM-3:55 PM
Room:250 F
For Part 2 see MS239 We address recent advances in algorithms for variational methods in imaging with a particular focus on non-rigid image registration. We also discuss the use of priors in order to supervise algorithms and provide more plausible results in particular applications. The associated largescale mathematical problems manifest as formidable, often ill-behaved systems. Key algorithmic challenges include computational complexity, memory consumption, a vast number unknowns as well as model and data uncertainties. We showcase state-of-the-art techniques in scientific computing to tackle these challenges.
Room:250 E For Part 3 see MS188 For Part 5 see MS237 Time-stepping methods are numerical methods for the time evolution of ODEs, DAEs, and PDEs in the method-of-lines framework. Many applications require specialized time-stepping methods in order to enhance efficiency or capture particular theoretical properties such as positivity, symplecticity, and strongstability-preservation or practical properties such as parallelization on emerging highperformance architectures, component (e.g., multirate or local time-stepping) and physics (IMEX or semi-implicit) splittings based on sub-system properties, and reliable error estimation and control. The talks in this minisymposium will describe recent novel developments in time-stepping methods and demonstrate their benefits in practical applications.
Organizer: Michael Guenther
For Part 1 see MS189 For Part 3 see MS238 By focusing effort on identifying dominant terms of a function representation one can often accurately approximate the function from limited data. Sparsity detection also referred to as compressed sensing and l1-minimization is one approach that is extremely effective at identifying the dominant terms of a function expansion. Recently there have been many advances in l1-minimization including strategies for sampling, preconditioning and highdimensional basis selection that attempt to maximize the accuracy of recovered signals whilst minimizing the number of signal samples. This minisymposium presents recent theoretical results, numerical methods and applications of sparse approximations, including but not limited to uncertainty quantification (UQ) and sensitivity analysis (SA).
Bergische Universität, Germany
Organizer: John D. Jakeman
Organizer: Adrian Sandu
Sandia National Laboratories, USA
Virginia Tech, USA
Organizer: Khachik Sargsyan
Organizer: Raymond J. Spiteri
Sandia National Laboratories, USA
Organizer: Emil M. Constantinescu Argonne National Laboratory, USA
University of Saskatchewan, Canada 2:15-2:35 Strong Stability Preserving General Linear Methods Zdzislaw Jackiewicz, Arizona State University, USA 2:40-3:00 Stability-Optimized Time Integrators for WENO Discretizations David I. Ketcheson, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Debojyoti Ghosh, Argonne National Laboratory, USA 3:05-3:25 Implicit-Explicit General Linear Methods Hong Zhang, Virginia Tech, USA 3:30-3:50 Optimal Explicit Strong Stability Preserving Runge–Kutta Methods with High Linear Order and Optimal Nonlinear Order Sigal Gottlieb, Zachary J. Grant, and Daniel L. Higgs, University of Massachusetts, Dartmouth, USA
2:15-2:35 Sampling Strategies for L1 Minimization John D. Jakeman, Sandia National Laboratories, USA; Akil Narayan, University of Massachusetts, Dartmouth, USA 2:40-3:00 Reweighted Minimization Method for Uncertainty Quantification of Microscopic Modeling Guang Lin, Xiu Yang, and Huan Lei, Pacific Northwest National Laboratory, USA 3:05-3:25 Least Square Methods for Low-Rank Approximations with Sparsity Inducing Regularization Prashant Rai, Ecole Centrale de Nantes, France; Mathilde Chevreuil, Université de Nantes, France; Loïc Giraldi and Anthony Nouy, Ecole Centrale de Nantes, France 3:30-3:50 Interpolation Via Weighted L1 Minimization Rachel Ward, University of Texas at Austin, USA; Holger Rauhut, RWTH Aachen University, Germany
Room:251 A
Organizer: Andreas Mang University of Texas at Austin, USA
Organizer: George Biros University of Texas at Austin, USA 2:15-2:35 Constrained Optimal Control Approaches in Large Deformation Diffeomorphic Metric Mapping Laurent Younes, Johns Hopkins University, USA 2:40-3:00 Nonlinear Image Registration with a Sliding Motion Deformation Model Alexander Derksen, University of Lübeck, Germany 3:05-3:25 Efficient Algorithms for HighResolution Diffusion-Weighted MRI Lars Ruthotto, Emory University, USA 3:30-3:50 Efficient Algorithms for Physically Constrained Diffeomorphic Image Registration Andreas Mang and George Biros, University of Texas at Austin, USA
2015 SIAM Conference on Computational Science and Engineering
135
Tuesday, March 17
Tuesday, March 17
Tuesday, March 17
MS215
MS216
MS217
Numerical Methods for High-Dimensional Stochastic and Parametric Problems - Part II of V 2:15 PM-3:55 PM
Water Resources Management: How to add it all up 2:15 PM-3:55 PM
Computational Benchmarks for Soft Materials and Complex Fluids at the Mesoscale - Part II of III 2:15 PM-3:55 PM
Room:251 B
Part of the SIAM Workshop Celebrating Diversity
For Part 1 see MS191 For Part 3 see MS240 Development of scalable numerical methods for the solution of problems with highdimensional stochastic or parametric inputs has been a subject of active research in computational sciences and engineering. This is motivated by the need to reduce the issue of curse-of-dimensionality, i.e., exponential increase of computational complexity, in predictive simulation of physical systems where accurate specification of governing laws entails a large number of parameters or stochastic variables. To this end, several novel approaches based on multi-level, reduced order, sparse, and lowrank approximations have been recently developed. This minisymposium presents state-of- the-art in such developments for various aspects of high-dimensional computation, including analysis, algorithms, implementation, and applications.
Organizer: Dongbin Xiu University of Utah, USA
Organizer: Alireza Doostan University of Colorado Boulder, USA 2:15-2:35 Sparse, Adaptive Smolyak Quadrature Algorithms for Stochastic Inverse Problems Christoph Schwab and Claudia Schillings, ETH Zürich, Switzerland 2:40-3:00 Accelerated Bayesian Inference with Transport Maps Matthew Parno and Youssef M. Marzouk, Massachusetts Institute of Technology, USA 3:05-3:25 Quasi Optimal Sparse-Grid Approximation of Random Elliptic PDEs Lorenzo Tamellini and Fabio Nobile, EPFL, Switzerland; Raul F. Tempone, King Abdullah University of Science & Technology (KAUST), Saudi Arabia 3:30-3:50 Title Not Available at Time of Publication Michael S. Eldred, Sandia National Laboratories, USA
Room:251 C
Techniques from systems theory are used to create decision support systems for the management of water resources. Development of sustainable engineering solutions in which the competing objectives of economic growth and environmental protection is aided by combined physics based models and optimization. These types of solutions have been found to be more effective when originating at a local level as it allows stakeholder/decision maker interaction. This session will present methods for managing limited and decreasing water resources at the local/ regional level.
Organizer: Elma A. Uddameri Texas Tech University, USA
Organizer: Raegan Higgins Texas Tech University, USA
Organizer: Susan E. Minkoff University of Texas at Dallas, USA
Organizer: Stephen Wirkus Arizona State University, USA 2:15-2:35 Approaches to Evaluate Interactions in Collaborative Groundwater Management Joseph Amaya, Texas A&M University, Kingsville, USA 2:40-3:00 Managing Surface Water Resources in Data Sparse Regions Felipe Estrada, Texas Tech University, USA 3:05-3:25 Application of SimulationOptimization for Water Management in Hydraulic Fracturing Operations Elma A. Uddameri, Texas Tech University, USA
Room:251 D For Part 1 see MS193 For Part 3 see MS242 Computational modeling of multiphysical processes at the mesoscale is becoming increasingly important for applications in micro-/nano-technology, next-generation battery technology, material assembly and biological systems. In this minisymposium we address mathematical modeling along with numerical schemes that can effectively capture mesoscale multiphysics such as hydrodynamics, transport, electrostatics and chemical reaction. Desirable features of the numerical schemes include robustness, accuracy, scalability, as well as potential to cross scales as a multiscaling framework. The benchmark applications that we are interested in include mixing and separation processes in micro-/nano-channel, semipermeable membranes, diffusive transport and electrochemical reaction in porous electrode, colloid and nanoparticle assembly, as well as macromolecular conformational change.
Organizer: Mauro Perego Sandia National Laboratories, USA
Organizer: Wenxiao Pan Pacific Northwest National Laboratory, USA
Organizer: Nathaniel Trask Brown University, USA
Organizer: Kyungjoo Kim Sandia National Laboratories, USA 2:15-2:35 Meshless Methods for the Mesoscale - High Order Implicit ALE Schemes using Collocated MLS Nathaniel Trask, Brown University, USA; Kyungjoo Kim and Mauro Perego, Sandia National Laboratories, USA
3:30-3:50 Climate Change and Water Scarcity Donna Mitchell, Texas Tech University, USA
continued on next page
136
Tuesday, March 17
MS217 Computational Benchmarks for Soft Materials and Complex Fluids at the Mesoscale - Part II of III 2:15 PM-3:55 PM continued
2:40-3:00 Efficient Parallel Implementation of Implicit SPH/MLS using LAMMPS and Trilinos Kyungjoo Kim and Mauro Perego, Sandia National Laboratories, USA; Nathaniel Trask, Brown University, USA; Michael L. Parks, Sandia National Laboratories, USA 3:05-3:25 Classical Density Functional Theory of Charged Fluids at Interfaces Amalie Frischknecht, Sandia National Laboratories, USA 3:30-3:50 Interface Resolved Numerical Method to Study Electrokinetic Particle Assembly in Microdevices Prashanta Dutta, Washington State University, USA
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS218 Inverse Problems and Data Assimilation - Part III of VI 2:15 PM-3:55 PM Room:251 E For Part 2 see MS194 For Part 4 see MS243 The aim of this minisymposium is to document recent mathematical developments in the field of inverse problems and data assimilation based on theoretical and numerical grounds that are relevant for various scientific and real life applications. Topics for the minisymposium include but are not limited to: 1) deterministic and statistical approaches to inversion problems and data assimilation; 2) reduced order modeling inversion framework; 3) treatment of nonlinear and non-smooth processes; 4) sensitivity analysis and its application to adaptive (targeting) observation; 5) operational data assimilation systems; 6) uncertainties impact studies; 7) automatic tools to support inversion and data assimilation methodologies.
Organizer: Razvan Stefanescu Virginia Tech, USA
Organizer: Adrian Sandu Virginia Tech, USA
Organizer: Ionel M. Navon Florida State University, USA
Organizer: Dacian N. Daescu Portland State University, USA 2:15-2:35 Bayesian Nonlinear Smoothing and Adaptive Sampling Pierre F.J Lermusiaux and Tapovan Lolla, Massachusetts Institute of Technology, USA 2:40-3:00 An Information Theoretic Approach to Use High-Fidelity Codes to Calibrate Low-Fidelity Codes Allison Lewis and Ralph C. Smith, North Carolina State University, USA; Brian Williams, Los Alamos National Laboratory, USA
continued in next column
3:05-3:25 Displacement Data Assimilation Juan M. Restrepo, Oregon State University, USA; Steven Rosenthal and Shankar C. Venkataramani, University of Arizona, USA; Arthur Mariano, University of Miami, USA 3:30-3:50 A Sampling Filter for NonGaussian Data Assimilation Ahmed Attia and Adrian Sandu, Virginia Tech, USA
2015 SIAM Conference on Computational Science and Engineering
137
Tuesday, March 17
Tuesday, March 17
Tuesday, March 17
MS219
MS220
MS221
Uncertainty Quantification Methods for Power Grid Systems - Part II of II 2:15 PM-3:05 PM
Numerical Methods for Multimaterial Flows with Deforming Boundaries Part II of III 2:15 PM-3:55 PM
Parallel Computing for Partial Differential Equations on CPUs, GPUs, and Intel Phi Part II of II 2:15 PM-3:55 PM
Room:254 A
Room:254 B
For Part 1 see MS196 For Part 3 see MS245 Multimaterial flows with deforming boundaries appear in many applications involving biological flows, medicine, industry, and homeland security. As the number of applications grow, so too the amount of research into developing and improving existing numerical methods for deforming boundary problems in multi-material flows. In this minisymposium, novel numerical methods will be presented that address efficiency and accuracy in adaptive mesh refinement algorithms, scalability on high performance computing platforms, robust interface capturing methods, numerical methods for phase change (freezing, boiling, evaporation), and numerical methods for enforcing boundary conditions at deforming boundary interfaces.
For Part 1 see MS197 Modern architectures including multi-core CPUs, massively parallel GPUs, and manycore accelerators such as the 60-core Intel Phi processor offer great opportunities for speeding up computations. They pose significant challenges due to the hybrid programming models and libraries that need to be used, ranging from distributed-memory computing with MPI, multi-threading with OpenMP, CUDA on GPUs, offloading to the Phi, for instance. This minisymposium will share experiences with parallel code and numerical algorithms from fundamental test problems over established benchmark codes to specialized research code. The minisymposium will start with an overview and contrast of the architectures and include the opportunity for the audience to discuss their own experiences.
Organizer: Mark Sussman Florida State University, USA
University of Maryland, Baltimore County, USA
Organizer: Frederic G. Gibou
Organizer: Ulrich J. Ruede
University of California, Santa Barbara, USA
University of Erlangen-Nuremberg, Germany
Organizer: Feng Xiao
2:15-2:35 Multigrid Solvers on Heterogeneous Architectures Björn Gmeiner, Daniel Iuhasz, Sebastian Kuckuk, Markus Stuermer, Harald Koestler, and Ulrich J. Ruede, University of Erlangen-Nuremberg, Germany
Room:251 F For Part 1 see MS195 The properties of many operational and structural elements of power grid models are typically known to some degree of certainty. Examples include grid topology and state, including knowledge of what lines are down; model parameters, such as generator/load/line electrical and performance characteristics; and operational conditions, such as renewable generator and load levels. In this minisymposium we bring together modeling contributions for the treatment of these uncertainties, to improve the design and operation of power grids.
Organizer: Cosmin Safta Sandia National Laboratories, USA
Organizer: Ali Pinar Sandia National Laboratories, USA
Organizer: Jean-Paul Watson Sandia National Laboratories, USA 2:15-2:35 Statistical Metrics for Assessing Quality of Scenarios for Unit Commitment and Dispatch Sarah M. Ryan and Didem Sari, Iowa State University, USA 2:40-3:00 Adaptive Robust Optimization with Dynamic Uncertainty Sets for Power System Operations Andy Sun and Alvaro Lorca, Georgia Institute of Technology, USA
Tokyo Institute of Technology, Japan 2:15-2:35 A Robust and Efficient Solver for Interfacial Multi-phase Flows on Unstructured Grids Feng Xiao, Bin Xie, and Sun Ziyao, Tokyo Institute of Technology, Japan 2:40-3:00 Reconstructed Discontinuous Galerkin (RDG) Method for MultiMaterial Flows on Unstructured Meshes Robert Nourgaliev and Sam Schofield, Lawrence Livermore National Laboratory, USA 3:05-3:25 A New Incompressibility Discretization for a Hybrid Particle Mac Grid Representation with Surface Tension Wen Zheng and Bo Zhu, Stanford University, USA; Byungmoon Kim, Adobe Systems, USA; Ronald Fedkiw, Stanford University, USA 3:30-3:50 An Eulerian Projection Method for Quasi-Static Elastoplasticity Chris H. Rycroft, Harvard University, USA
Organizer: Matthias K. Gobbert
2:40-3:00 Speeding Up Sparse Triangular Solution on Multicores and GPUs Humayun Kabir, Joshua D. Booth, and Padma Raghavan, Pennsylvania State University, USA 3:05-3:25 Applications of Distributed Methods to Non-Traditional Linear Systems Julian Gilyard, Wake Forest University, USA; Thomas Stitt, Pennsylvania State University, USA; Oluwapelumi Adenikinju and Joshua Massey, University of Maryland, Baltimore County, USA 3:30-3:50 General SpMV and SpMM for AMG on GPUs Robert Strzodka, NVIDIA, USA
138
Tuesday, March 17
MS222 Industrial Mathematics Education 2:15 PM-3:55 PM Room:254 C Organized by the SIAM Education Committee Industrial Mathematics Education takes many forms: internships, mathematical modeling courses, team projects and research collaborations. This minisymposium will present industrial mathematics from multiple perspectives, such as faculty mentor, industry liaison and student researcher.
Organizer: Rachel Levy Harvey Mudd College, USA 2:15-2:35 Industrial Mathematics Education at Worcester Polytechnic Institute Marcel Blais, Worcester Polytechnic Institute, USA 2:40-3:00 A Student Perspective on Industrial Capstones at Harvey Mudd College Elizabeth Schofield, Harvey Mudd College, USA 3:05-3:25 A New Curriculum in Applied and Computational Mathematics Jeffrey Humpherys, Brigham Young University, USA 3:30-3:50 PIC Math: Preparation for Industrial Careers in Mathematical Sciences Michael Dorff, Brigham Young University, USA
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
Tuesday, March 17
MS223
MS224
Hybrid and Multilevel Approaches to Kinetic Equations - Part IV of IV 2:15 PM-3:55 PM
Low-rank Optimization with Applications - Part I of II 2:15 PM-3:55 PM
Room:150 AB
Oak Ridge National Laboratory, USA
For Part 2 see MS249 Low-rank matrices and tensors are a popular modeling tool in machine learning and an effective approximation class in many highdimensional problems. At the same time, low-rank approximations can potentially lead to significant computational savings in algorithms that have to deal with large and dense matrices. This minisymposium showcases recent work in this area by focusing on low-rank optimization. The applications include subspace tracking, nonlinear programming, tensor completion, and high-dimensional eigenvalue problems, while the algorithmic techniques involve greedy algorithms, stochastic optimization, preconditioning, and optimization on manifolds.
Organizer: Ryan McClarren
Organizer: Bart Vandereycken
For Part 3 see MS199 In recent years, a significant amount of effort has been directed to solving kinetic equations. Generally, kinetic equations have posed a challenge due to the large phase space associated with the equations, however algorithmic advances and advances in computer hardware have made these simulations more tractable. In this minisymposium, we consider hybrid and multilevel approaches to solving kinetic equations.
Organizer: Jeffrey A. Willert Los Alamos National Laboratory, USA
Organizer: Cory Hauck
Room:150 DE
Texas A&M University, USA
Princeton University, USA
Organizer: Martin Frank
Organizer: Daniel Kressner
RWTH - Aachen University of Technology, Germany
Organizer: Jingmei Qiu University of Houston, USA 2:15-2:35 A Multigrid Method for Two-Dimensional Discrete-Ordinates Radiation-Transport Calculations Jeffery D. Densmore, Daniel Gill, and Justin Pounders, Bettis Laboratory, USA 2:40-3:00 Multilevel Projection Method for Nonlinear Radiative Transfer Problems Dmitriy Y. Anistratov, North Carolina State University, USA 3:05-3:25 Multilevel Monte Carlo Methods for Kinetic Equations Lee F. Ricketson, Courant Institute of Mathematical Sciences, New York University, USA 3:30-3:50 A Conservative High-Order/ Low-Order Method Based Upon a Non-Conservative High-Order Least Squares Sn Formulation Jacob Peterson and Jim E. Morel, Texas A&M University, USA
EPFL, Switzerland 2:15-2:35 Preconditioned Riemannian Optimization for Low-Rank Tensor Equations Bart Vandereycken, Princeton University, USA 2:40-3:00 Semi-Supervised Robust Matrix Completion for Dynamic Subspace Estimation and Tracking Hassan Mansour, Mitsubishi Electric Research Laboratories, USA 3:05-3:25 Low-rank Approximation of Matrices and Tensors for Dynamical and Optimization Problems Ivan Oseledets, Denis Kolesnikov, and Mikhail Litsarev, Skolkovo Institute of Science and Technology, Russia 3:30-3:50 Greedy Algorithms for Parametric Eigenvalue Problems Virginie Ehrlacher, CERMICS, France
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS225 Deployment and Application of Technologies provided by the FASTMath Institute - Part II of II 2:15 PM-3:55 PM Room:258 For Part 1 see MS201 The FASTMath SciDAC Institute develops and deploys scalable mathematical algorithms and software tools for reliable simulation of complex physical phenomena. These tools include structured and unstructured mesh techniques, linear and nonlinear solvers, eigensolvers and time integration methods. Research efforts are focused on developing a full range of accurate and robust technologies that support different application codes specifically those employed by the Department of Energy’s domain scientists. This minisymposium will place emphasis on application of FASTMath tools on SciDAC and related applications. Speakers will discuss the challenges in different application areas and the impact of the latest developments in FASTMath tools.
Organizer: Onkar Sahni Rensselaer Polytechnic Institute, USA
Organizer: Lori A. Diachin Lawrence Livermore National Laboratory, USA 2:15-2:35 Massively Parallel Flow Simulation using PETSc Michel Rasquin and Benjamin Matthews, University of Colorado Boulder, USA; Cameron Smith, Rensselaer Polytechnic Institute, USA; Kenneth Jansen, University of Colorado Boulder, USA 2:40-3:00 Albany: A Trilinos-based code for Ice Sheet Simulations and other Applications Andrew Salinger, Glen Hansen, Irina Kalashnikova, Mauro Perego, and Ray S. Tuminaro, Sandia National Laboratories, USA
continued in next column
3:05-3:25 Accurate Adaptive Loops for Finite Deformation Plasticity in Albany Brian Granzow, Rensselaer Polytechnic Institute, USA; Glen Hansen, Sandia National Laboratories, USA; Dan A. Ibanez and Mark S. Shephard, Rensselaer Polytechnic Institute, USA 3:30-3:50 Variational Multiscale Analysis of Stochastic Partial Differential Equations in Albany Onkar Sahni, Jason Li, Jayanth JagalurMohan, and Assad Oberai, Rensselaer Polytechnic Institute, USA
139
Tuesday, March 17
MS226 Advances in Multigrid Methods and Their Applications - Part II of II 2:15 PM-3:55 PM Room:259 For Part 1 see MS202 Multigrid is the most successful and generally-applicable scalable approach to solving problems with a hierarchy of scales in application areas such as elliptic PDEs, optimization, inversion, and image processing. Scalable methods are increasingly essential as hardware evolution and algorithmic advances enable the solution of larger scale problems in science and engineering applications. This session brings together researchers developing multigrid and related methods to assess the state of the art, recent theoretical advances, and novel applications.
Organizer: Zichao Di Argonne National Laboratory, USA
Organizer: Maria Emelianenko George Mason University, USA
Organizer: Jed Brown Argonne National Laboratory, USA and University of Colorado Boulder, USA 2:15-2:35 Fast Algorithms for Shape Analysis of Planar Objects Gunay Dogan, National Institute of Standards and Technology, USA 2:40-3:00 Deflation-based Domain Decomposition Methods Pierre Jolivet, ETH Zürich, Switzerland; Frederic Nataf, Laboratoire Jacques-Louis Lions, France; Christophe Prud’homme, University of Strasbourg, France 3:05-3:25 The Auxiliary Space Solvers and Its Applications Lu Wang, Pennsylvania State University, USA 3:30-3:50 Multiscale Methods for Networks Ilya Safro, Clemson University, USA
140
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
Tuesday, March 17
MS228
MS229
Recent Advances in High Order Spatial Discretization Methods for PDEs Part II of III 2:15 PM-3:55 PM
Efficient Solvers for PDEconstrained Optimization Part I of II 2:15 PM-3:55 PM
For Part 3 see MS203 Due to recent advances in the underlying technology, integral equations have become highly competitive in attacking the most challenging of elliptic, globally-coupled PDE problems. However, compared to a standard off-the-shelf finite element methods, a larger amount of machinery is involved, including fast algorithms, preconditioners, and singular quadrature methods. To make this theoretically attractive family of methods practically usable, many of these components need to be readily available in software form. This minisymposium seeks to bring together practitioners and researchers with the goal of encouraging exchange of ideas and discovering opportunities for collaboration in this direction.
Room:260 B
For Part 2 see MS252 Optimization problems constrained by partial differential equations arise in a wide range of applications in science and engineering. From a computational viewpoint their solution pose various challenges. One of these challenges is the development of efficient solution methods, which must often be adapted to each specific problem class. The aim of this minisymposium is to bring together researchers to report on recent developments on efficient methods for solving PDE-constrained problems.
Organizer: Timo Betcke
Tuesday, March 17
MS227 Software Components for Integral Equation Methods Part IV of IV 2:15 PM-3:55 PM Room:260 A
For Part 1 see MS204 For Part 3 see MS251 Many complex physical phenomena can be modeled by partial differential equations and the numerical simulation of PDEs has been an active research area. This minisymposium aims to bring researchers together to discuss the numerical challenges when designing high order spatial discretization methods for time-dependent/static PDEs, with particular attention on weighted essentially non-oscillatory (WENO) methods and discontinuous Galerkin (DG) methods.
Room:151 AB
Organizer: Simon W. Funke
Organizer: Yuan Liu
Simula Research Laboratory, Norway
Michigan State University, USA
Organizer: Kent-Andre Mardal
Organizer: Qi Tang
University College London, United Kingdom
Michigan State University, USA
University of Oslo, and Simula Research Laboratory, Norway
Organizer: Andreas Kloeckner
Organizer: Andrew J. Christlieb
Organizer: Marie E. Rognes
Michigan State University, USA
Simula Research Laboratory, Norway
2:15-2:35 Robust Algorithms for Periodic Problems and Evaluation of Layer Potentials Alex H. Barnett, Dartmouth College and Simons Foundation, USA; Shravan Veerapaneni, University of Michigan, USA; Adrianna Gillman, Rice University, USA; Min Hyung Cho, Lin Zhao, and Yuxiang Liu, Dartmouth College, USA; Bowei Wu and Gary Marple, University of Michigan, USA; Leslie Greengard, Simons Foundation and Courant Institute of Mathematical Sciences, New York University, USA
2:15-2:35 A Simple Weighted Essentially Non-oscillatory Limiter for the Correction Procedure via Reconstruction (CPR) Framework Jie DU Jie, University of Science and Technology of China, China; Chi-Wang Shu, Brown University, USA; Mengping Zhang, University of Science and Technology of China, China
2:15-2:35 Nonstandard Sobolev Spaces for Preconditioning Mixed Methods Walter Zulehner, University of Linz, Austria; Wolfgang Krendl, Johannes Kepler University Linz, Austria
2:40-3:00 Title Not Available at Time of Publication Denis Zorin, Courant Institute of Mathematical Sciences, New York University, USA
3:05-3:25 A Compact-Reconstruction WENO Scheme with Semi-Implicit Time Integration Debojyoti Ghosh and Emil M. Constantinescu, Argonne National Laboratory, USA
University of Illinois, USA
3:05-3:25 Adaptive Boundary Element Methods Michael Feischl, Thomas Führer, Gregor Ganter, Alexander Haberl, and Dirk Praetorius, Vienna University of Technology, Austria 3:30-3:50 Fast Algorithms for the Evaluation of Layer Potentials using ‘Quadrature by Expansion’ Andreas Kloeckner, University of Illinois, USA
2:40-3:00 Compact-Reconstruction WENO on Non-uniform Meshes Kilian Cooley and James Baeder, University of Maryland, USA
3:30-3:50 Superconvergence Properties of Discontinuous Galerkin Methods Based on Upwind-Biased Fluxes for Linear Hyperbolic Equations Daniel Frean, University of East Anglia, United Kingdom
2:40-3:00 Preconditioning of Active-Set Newton Methods for PDE-Constrained Optimal Control Problems Margherita Porcelli, Valeria Simoncini, and Mattia Tani, Universita’ di Bologna, Italy 3:05-3:25 HPC Methods for Structured Inverse Modeling in Diffusive Processes Martin Siebenborn and Volker H. Schulz, University of Trier, Germany 3:30-3:50 Parallel Preconditioning for Time-Dependent PDE-Constrained Optimization Eleanor McDonald and Andy Wathen, University of Oxford, United Kingdom
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS230 Advances in Computational Techniques for Coastal Ocean Modeling Part II of II 2:15 PM-3:55 PM Room:151 DE For Part 1 see MS206 Environmental and coastal ocean engineering models yield complex systems combining interdisciplinary techniques. Moreover, their accurate and efficient simulation requires advanced techniques in high performance scientific computing. This minisymposium brings together experts to discuss a number of new ideas for the development, analysis and application of this important class of problems. Specific ideas include the high resolution modeling of weather extremes such as hurricane storm surge, wave propogation, climate change, and flood control.
Organizer: Jennifer Proft University of Texas at Austin, USA
Organizer: Clint Dawson University of Texas at Austin, USA
Tuesday, March 17
Tuesday, March 17
MS231
MT4
Advances in High Level Finite Element Systems - Part II of II 2:15 PM-3:55 PM
Minitutorial: Lab Skills for Scientific Computing Part II of II 4:25 PM-6:05 PM
Room:151 G For Part 1 see MS207 Many modern software projects utilize the high level of mathematical abstraction provided by finite element methods to develop powerful libraries that hide or even automate many practical computational details. Such codes increase programmer productivity, decrease potential sources of errors, and simplify the implementation of complex calculations such as h/p adaptivity and adjoint-based optimization. This minisymposium will bring together key participants from Deal.II, Dune, FEniCS, Dolfin-adjoint, Firedrake and Nektar++ to communicate the latest advances in and applications of such finite element systems.
Organizer: David Ham Imperial College London, United Kingdom
Organizer: Rob Kirby Baylor University, USA
2:15-2:35 Computational Modeling of Storm Surge in Galveston Bay Jennifer Proft, University of Texas at Austin, USA
2:15-2:35 Mesh-Independent Convergence for PDE-Constrained Optimisation Solvers in Dolfin-Adjoint Simon W. Funke and Magne Nordaas, Simula Research Laboratory, Norway
2:40-3:00 Three-Dimensional Coupled Wind-Wave and Cohesive Sediment Transport Modeling in South San Francisco Bay Oliver Fringer, Stanford University, USA
2:40-3:00 Simulating Coupled Pressure-Temperature Equations for Trace Gas Sensors Using FEniCS and PETSc Brian W. Brennan, Baylor University, USA
3:05-3:25 Strengthening the Hurricane Wave and Surge Forecast Guidance provided to Coastal Communities in North Carolina Rosemary Cyriac and J. Casey Dietrich, North Carolina State University, USA; Jason Fleming, Seahorse Coastal Consulting, USA; Brian Blanton, University of North Carolina, Chapel Hill, USA; Rick Luettich, University of North Carolina, Moorehead City, USA
3:05-3:25 Spectral/HP Element Modelling in Nektar++ David Moxey, Chris Cantwell, and Spencer Sherwin, Imperial College London, United Kingdom; Mike Kirby, University of Utah, USA
3:30-3:50 Discontinuous Galerkin Methods for Spectral Wave/Circulation Modeling Jessica Meixner, University of Notre Dame, USA
141
3:30-3:50 Supporting Modern HPC Hardware in the DUNE Framework Christian Engwer and Fahlke Jorrit, University of Münster, Germany; Steffen Müthing, Heidelberg University, Germany
Coffee Break 3:55 PM-4:25 PM Room:255
Room:355 For Part 1 see MT3 Chair: Greg Wilson, Mozilla Foundation, USA The Software Carpentry project (http:// software-carpentry.org) has been teaching basic computing skills to scientists and engineers since 1998. This minitutorial will introduce the tools and techniques that have proven most useful, and show how integrating them can help researchers get more done in less time, and with less pain. This two-part minitutorial will introduce several widely-used practices in software development, explore the empirical evidence showing their benefits (or in some cases the lack thereof), and describe how researchers and research teams can adopt them. Some of the work will be hands-on, so participants are strongly urged to bring a laptop. Warning: real-world examples may be used. Greg Wilson, Mozilla Foundation, USA
142
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
Tuesday, March 17
MS167
MS232
MS233
Reconnection-based Arbitrary LagrangianEulerian Methods 4:25 PM-6:05 PM
Educational Applications of Agent-Based Modeling 4:25 PM-6:05 PM Room:355 A
Room:259
This minisymposium will focus on educational applications of agent-based simulations across science and engineering curricula at all levels with a variety of technology environments. Specialized tools such as AgentSheets and NetLogo, along with Excel and JavaScript templates, introduce students to agent-based modeling, computational thinking, and their underlying mathematical foundation. College faculty and high school teachers will demonstrate how interesting agent-based simulations inform students about the content and method of numerical modeling while increasing the number and diversity of students excited about science. Moreover, agent-based modeling enhances student understanding of computational science and appreciation of its rich variety of scientific applications.
Recent Advances in Numerical Methods for Interface Problems Part II of IV 4:25 PM-6:05 PM
Tuesday, March 17
In numerical simulations of multidimensional fluid flow, the relationship of the motion of the computational grid to the motion of the fluid is an important issue. The motion of the grid can be chosen arbitrarily. The philosophy of the arbitrary Lagrangian-Eulerian methodology (ALE) is to exploit this degree of freedom to improve the accuracy and efficiency of the simulation. In talks presented at this minisymposium new reconnection-based ALE method in which mesh allowed to change connectivity is explored.
Organizer: Mikhail Shashkov Los Alamos National Laboratory, USA
Organizer: J. Michael Owen Lawrence Livermore National Laboratory, USA 4:25-4:45 Adaptive ReconnectionBased Arbitrary Lagrangian Eulerian Method Wurigen Bo and Misha Shashkov, Los Alamos National Laboratory, USA 4:50-5:10 Reconnection ALE in a Massively-Parallel, Staggered-Grid, Multi-Physics Code David Starinshak, J. Michael Owen, and Douglas S. Miller, Lawrence Livermore National Laboratory, USA 5:15-5:35 Multimaterial Simulation in Reale Framework Jerome Breil, University of Bordeaux, France 5:40-6:00 Triangular Metric-Based Mesh Adaptation for Compressible Multi-Material Flows in SemiLagrangian Coordinates Stephane Del Pino, CEA Saclay, France; Isabelle Marmajou, CEA, France
Organizer: Angela B. Shiflet Wofford College, USA
Organizer: Robert M. Panoff Shodor, USA 4:25-4:45 Applying Run-Modify-Build Templates for Agent-Based Models Robert M. Panoff, Shodor, USA 4:50-5:10 Teaching Freshman Science Using Agent-Based Computational Laboratories George W. Shiflet and Angela B. Shiflet, Wofford College, USA 5:15-5:35 Transitioning from Game Design to Simulation Using AgentBased Modeling Fred Gluck, University of Colorado Boulder, USA 5:40-6:00 NetLogo in the Secondary Life Science Classroom Charlotte M. Trout, Retired
Room:250 A For Part 1 see MS208 For Part 3 see MS257 Many real world applications are characterized by multiple materials and complex fluids, and they often lead to interface problems. It is imperative to develop efficient and stable numerical methods for these problems. Great efforts have been made for solving interface problems and tracing the moving interfaces in the past decades. However, many challenges, such as the lack of regularity of physical solutions due to the discontinuity across interfaces, still hinder the development of efficient numerical methods. This minisymposium intends to create a forum for researchers from different fields to discuss recent advances on the interface methods and their applications.
Organizer: Xiaoming He Missouri University of Science and Technology, USA
Organizer: Xiaolin Li State University of New York, Stony Brook, USA 4:25-4:45 Microstructure for Free Surface Flows James G. Glimm, State University of New York, Stony Brook, USA 4:50-5:10 A Fictitious Domain Method with a Hybrid Cell Model for Simulating Motion of Cells in Fluid Flow Zhiliang Xu, University of Notre Dame, USA 5:15-5:35 Volume-Preserving Adaptive Moment-of-Fluid Method for Interface Tracking Shengtai Li, Los Alamos National Laboratory, USA; Hyung T. Ahn, University of Ulsan, South Korea; Mikhail Shashkov, Los Alamos National Laboratory, USA 5:40-6:00 Dissipation and Dispersion Errors of Discontinuous Galerkin Method and Its Application to Level Set Equations Jue Yan, Iowa State University, USA
2015 SIAM Conference on Computational Science and Engineering
143
Tuesday, March 17
Tuesday, March 17
Tuesday, March 17
MS234
MS235
MS236
First Principle Methods and Applications for Computational Materials Science and Chemistry Part III of III 4:25 PM-6:05 PM
Enhancing Software Development for Emerging Platforms using Algorithms and Performance Tools 4:25 PM-6:05 PM
Recent Advances in Model Reduction - Part III of V 4:25 PM-6:05 PM
Room:250 B
Software development needs urgent help to keep up with the rapidly emerging hardware platforms. With heterogeneity in these platforms, exploiting hybrid parallelism has added on to the already existing complexity of software development. Newer algorithms, auto-tuning and load balancing techniques, and much more are required to exploit the multi-level concurrency. In this MS, we will discuss selection of algorithms and scheduling techniques for accelerator-based codes for numerical linear algebra, the role of performance tools to assist porting applications to newer platforms, low-power algorithms and its adaptation on Tilera and experimental analysis demonstrating methodologies employed to effectively leverage heterogeneous memory subsystems.
For Part 2 see MS209 First principle electronic structure calculations are among the most challenging and computationally demanding science and engineering problems. This minisymposium aims at presenting and discussing new numerical methods for achieving ever higher level of accuracy and efficiency in ground and excited density functional theory calculations, and their applications in computational materials science and chemistry. We propose to bring together physicists, chemists and materials scientists who are involved in improving the numerical development of widely known quantum chemistry and solidstate physics application software packages, with mathematicians and computer scientists who are focusing on advancing the required state-of-the-art mathematical algorithms and parallel implementation.
Organizer: Lin Lin University of California, Berkeley, USA
Organizer: Eric Polizzi University of Massachusetts, Amherst, USA
Organizer: Chao Yang Lawrence Berkeley National Laboratory, USA 4:25-4:45 A Parallel Orbital-Updating Approach for Electronic Structure Calculations Based on Singularity Decompositions Aihui Zhou, Chinese Academy of Sciences, China 4:50-5:10 Recent Progress on Quantum Mechanics Embedding Theory Chen Huang, Florida State University, USA 5:15-5:35 Towards Predictive Modeling of Correlation Effects in Many-electron Systems Karol Kowalski, Pacific Northwest National Laboratory, USA 5:40-6:00 Massively Parallel GW Calculations for Current and Nextgeneration HPC Jack Deslippe, National Energy Research Scientific Computing Center, USA
Room:250 C
Organizer: Sunita Chandrasekaran University of Houston, USA
Organizer: Barbara Chapman University of Houston, USA 4:25-4:45 Algorithmic Selection, Autotuning, and Scheduling for Accelerator-Based Codes for Numerical Linear Algebra Piotr Luszczek, University of Tennessee, Knoxville, USA 4:50-5:10 Using Performance Tools to Assist Porting to New Platforms Guido Juckeland, TU Dresden, Germany 5:15-5:35 Toward Heterogeneous Memory Systems for HPC Antonio J. Peña, Argonne National Laboratory, USA 5:40-6:00 Algorithmic Adaptations for Scalable Community Detection on the Tilera Many-Core Architecture Daniel Chavarria, Howard Lu, and Mahantesh Halappanavar, Pacific Northwest National Laboratory, USA; Ananth Kalyanaraman, Washington State University, USA
Room:250 D For Part 2 see MS211 For Part 4 see MS260 Model reduction has become an increasingly important tool to mitigate the computational burden of modeling and simulation in timecritical (e.g., model-predictive control) and many query (e.g., Bayesian inversion) applications. This minisymposium presents recent advances that address the primary challenges facing such methods, such as preserving intrinsic problem structure, handling high- dimensional parameter spaces, integration with existing simulation codes, stability, optimization and uncertainty quantification and control.
Organizer: Kevin T. Carlberg Sandia National Laboratories, USA
Organizer: Gianluigi Rozza SISSA, International School for Advanced Studies, Trieste, Italy 4:25-4:45 pyMOR - A New Model Order Reduction Software Framework Rene Milk, Mario Ohlberger, Stephan Rave, and Felix Schindler, University of Muenster, Germany 4:50-5:10 Reduced Basis Methods for Option Pricing Julien Salomon, CEREMADE Universite Paris 9 Dauphine, France; Olena Burkovska, Technical University of Munich, Germany; Bernard Haasdonk, University of Stuttgart, Germany; Barbara Wohlmuth, Technical University of Munich, Germany 5:15-5:35 Adaptivity and Reduced Basis Methods Karsten Urban, University of Ulm, Germany 5:40-6:00 Reduced Order Models for Patient-Specific Haemodynamics of Coronary Artery Bypass Grafts Francesco Ballarin and Elena Faggiano, Politecnico di Milano, Italy; Sonia Ippolito, Luigi Sacco Hospital, Italy; Andrea Manzoni and Alfio Quarteroni, École Polytechnique Fédérale de Lausanne, Switzerland; Gianluigi Rozza, SISSA, International School for Advanced Studies, Trieste, Italy; Roberto Scrofani, Luigi Sacco Hospital, Italy
144
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS237 Advances in Time-stepping Methods - Part V of V 4:25 PM-6:05 PM Room:250 E For Part 4 see MS212 Time-stepping methods are numerical methods for the time evolution of ODEs, DAEs, and PDEs in the method-of-lines framework. Many applications require specialized time-stepping methods in order to enhance efficiency or capture particular theoretical properties such as positivity, symplecticity, and strongstability-preservation or practical properties such as parallelization on emerging highperformance architectures, component (e.g., multirate or local time-stepping) and physics (IMEX or semi-implicit) splittings based on sub-system properties, and reliable error estimation and control. The talks in this minisymposium will describe recent novel developments in time-stepping methods and demonstrate their benefits in practical applications.
Organizer: Emil M. Constantinescu Argonne National Laboratory, USA
Organizer: Michael Guenther Bergische Universität, Germany
Organizer: Adrian Sandu Virginia Tech, USA
Organizer: Raymond J. Spiteri University of Saskatchewan, Canada 4:25-4:45 A Massively Parallel Solver for the Incompressible Navier--Stokes Equations Raymond J. Spiteri, University of Saskatchewan, Canada 4:50-5:10 Developing a Custom Time Integrator for the Nonlinear Schrödinger Equation for An Application in Paraxial Laser Propagation Michael Minion, Lawrence Berkeley National Laboratory, USA
continued in next column
5:15-5:35 Spatially Partitioned Embedded Runge-Kutta Methods Steven Ruuth, Simon Fraser University, Canada; David I. Ketcheson, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Colin B. Macdonald, Oxford University, United Kingdom 5:40-6:00 An Accelerated Domain Decomposition Method for Time Dependent Problems Ronald Haynes, Memorial University, Newfoundland, Canada
Tuesday, March 17
MS238 Sparse and Compressible Representations: Theory, Algorithms and Applications - Part III of III 4:25 PM-6:05 PM Room:250 F For Part 2 see MS213 By focusing effort on identifying dominant terms of a function representation one can often accurately approximate the function from limited data. Sparsity detection also referred to as compressed sensing and l1-minimization is one approach that is extremely effective at identifying the dominant terms of a function expansion. Recently there have been many advances in l1-minimization including strategies for sampling, preconditioning and high-dimensional basis selection that attempt to maximize the accuracy of recovered signals whilst minimizing the number of signal samples. This minisymposium presents recent theoretical results, numerical methods and applications of sparse approximations, including but not limited to uncertainty quantification (UQ) and sensitivity analysis (SA).
Organizer: John D. Jakeman Sandia National Laboratories, USA
Organizer: Khachik Sargsyan Sandia National Laboratories, USA 4:25-4:45 Coherence Motivated Sampling of Polynomial Chaos Expansions Jerrad Hampton and Alireza Doostan, University of Colorado Boulder, USA 4:50-5:10 Sparse Solutions to LargeScale Nonlinear Subsurface Flow Inverse Problems Benham Jafarpour and Reza Khaninezhad, University of Southern California, USA 5:15-5:35 An Efficient Method for the Computation of the Stochastic Galerkin Projections by Means of Tensor Format Representations Alexander Litvinenko, King Abdullah University of Science & Technology (KAUST), Saudi Arabia; Mike Espig, RWTH - Aachen University of Technology, Germany; Matthies Hermann G., TU Braunschweig, Germany; W. Hackbusch, Max Planck Institute for Mathematics in the Sciences, Germany 5:40-6:00 Tensor Approximation Methods for Stochastic Problems Elmar Zander, Technical University Braunschweig, Germany
2015 SIAM Conference on Computational Science and Engineering
145
Tuesday, March 17
Tuesday, March 17
Tuesday, March 17
MS239
MS240
MS241
Efficient Algorithms for Variational Methods in Imaging - Part II of II 4:25 PM-6:05 PM
Numerical Methods for High-Dimensional Stochastic and Parametric Problems - Part III of V 4:25 PM-6:05 PM
Linear-Complexity Dense Linear Algebra, Parallelization and Applications 4:25 PM-6:05 PM
Room:251 B
Room:251 C
For Part 2 see MS215 For Part 4 see MS264 Development of scalable numerical methods for the solution of problems with highdimensional stochastic or parametric inputs has been a subject of active research in computational sciences and engineering. This is motivated by the need to reduce the issue of curse-of-dimensionality, i.e., exponential increase of computational complexity, in predictive simulation of physical systems where accurate specification of governing laws entails a large number of parameters or stochastic variables. To this end, several novel approaches based on multi-level, reduced order, sparse, and low-rank approximations have been recently developed. This minisymposium presents state-of- the-art in such developments for various aspects of high-dimensional computation, including analysis, algorithms, implementation, and applications.
Many applications in such as quantum chemistry and seismic imaging give rise to dense operators. Well-known fast methods (e.g., FMM) for applying or inverting the operators admit an algebraically orientated interpretation. Such an interpretation inspires the study of matrices with a hierarchical low-rank structure, which enables linearcost algorithms for performing matrix operations. These operations include matrix-vector multiplication, inversion, determinant calculation and several others. This minisymposium presents latest advances in the matrix algorithms and their parallelization.
Room:251 A For Part 1 see MS214 We address recent advances in algorithms for variational methods in imaging with a particular focus on non-rigid image registration. We also discuss the use of priors in order to supervise algorithms and provide more plausible results in particular applications. The associated largescale mathematical problems manifest as formidable, often ill-behaved systems. Key algorithmic challenges include computational complexity, memory consumption, a vast number unknowns as well as model and data uncertainties. We showcase state-of-the-art techniques in scientific computing to tackle these challenges.
Organizer: Andreas Mang University of Texas at Austin, USA
Organizer: George Biros University of Texas at Austin, USA 4:25-4:45 Platform-independent Description of Image Registration Algorithms Harald Koestler, University of ErlangenNuremberg, Germany 4:50-5:10 Novel Priors and Algorithms for 4d Tracking and Classification of Cells Christoph Brune, University of Twente, Netherlands 5:15-5:35 Incorporating Uncertainty in MR Images of Glioblastoma when Leveraging Models to Interpret Therapeutic Efficacy Andrea Hawkins-Daarud and Kristin R. Swanson, Northwestern University, USA 5:40-6:00 Parameter Estimation for Malignant Brain Tumors Amir Gholaminejad and George Biros, University of Texas at Austin, USA
Organizer: Alireza Doostan University of Colorado Boulder, USA
Organizer: Dongbin Xiu University of Utah, USA 4:25-4:45 QoI Basis Adaptation Roger Ghanem, University of Southern California, USA; Kenny Chowdhary and Habib N. Najm, Sandia National Laboratories, USA 4:50-5:10 Numerical Methods for SPDEs with Levy Jump Processes SPDES: Stochastic and Deterministic Approaches Mengdi Zheng and George E. Karniadakis, Brown University, USA 5:15-5:35 Optimal Least-Squares Projection: Applications to Uq Tao Zhou, Chinese Academy of Sciences, China; Akil Narayan, University of Massachusetts, Dartmouth, USA; Dongbin Xiu, University of Utah, USA 5:40-6:00 Computational Complexity of Stochastic Galerkin and Collocation Methods for PDEs with Random Coefficients Nick Dexter, University of Tennessee, USA; Clayton G. Webster and Guannan Zhang, Oak Ridge National Laboratory, USA
Organizer: Jie Chen Argonne National Laboratory, USA 4:25-4:45 Linear-Cost Storage and Computation with Kernel Matrices Jie Chen, Argonne National Laboratory, USA 4:50-5:10 N-Body Algorithms for Matrices with Decay: Multiplication, Projection, Inverse Factorization & Fock-Exchange Matt Challacombe and Nicolas Bock, Los Alamos National Laboratory, USA; Terry Haut, University of Colorado Boulder, USA 5:15-5:35 Parallel Structured Direct Solvers for Nonsymmetric and Indefinite Sparse Matrices Zixing Xin, Purdue University, USA 5:40-6:00 Solvers for O(N) Electronic Structure in the Strong Scaling Limit Nicolas Bock, Los Alamos National Laboratory, USA; Terry Haut, University of Colorado Boulder, USA; Matt Challacombe, Los Alamos National Laboratory, USA; Laxmikant Kale, University of Illinois at UrbanaChampaign, USA
146
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS242 Computational Benchmarks for Soft Materials and Complex Fluids at the Mesoscale - Part III of III 4:25 PM-6:05 PM Room:251 D For Part 2 see MS217 Computational modeling of multiphysical processes at the mesoscale is becoming increasingly important for applications in micro-/nano-technology, next-generation battery technology, material assembly and biological systems. In this minisymposium we address mathematical modeling along with numerical schemes that can effectively capture mesoscale multiphysics such as hydrodynamics, transport, electrostatics and chemical reaction. Desirable features of the numerical schemes include robustness, accuracy, scalability, as well as potential to cross scales as a multiscaling framework. The benchmark applications that we are interested in include mixing and separation processes in micro-/nano-channel, semipermeable membranes, diffusive transport and electrochemical reaction in porous electrode, colloid and nanoparticle assembly, as well as macromolecular conformational change.
Organizer: Mauro Perego Sandia National Laboratories, USA
Organizer: Wenxiao Pan Pacific Northwest National Laboratory, USA
Organizer: Nathaniel Trask Brown University, USA
Organizer: Kyungjoo Kim Sandia National Laboratories, USA 4:25-4:45 Fluctuating Hydrodynamics Methods for Electrokinetics and Capillary Electrophoresis of Charged Colloids Paul J. Atzberger, University of California, Santa Barbara, USA 4:50-5:10 Biomechanics of Red Blood Cells and Related Diseases Zhangli Peng, University of Notre Dame, USA
continued in next column
5:15-5:35 Mesoscale Models for Molecular Solvation: Funny Business at the Solute-Solvent Interface Jaydeep Bardhan, Northeastern University, USA; Matthew G. Knepley, University of Chicago, USA 5:40-6:00 Mesoscopic Modeling of Temperature-dependent Properties in Non-isothermal Fluid Systems Zhen Li, Yu-Hang Tang, Bruce Caswell, and George E. Karniadakis, Brown University, USA
Tuesday, March 17
MS243 Inverse Problems and Data Assimilation - Part IV of VI 4:25 PM-6:05 PM Room:251 E For Part 3 see MS218 For Part 5 see MS267 The aim of this minisymposium is to document recent mathematical developments in the field of inverse problems and data assimilation based on theoretical and numerical grounds that are relevant for various scientific and real life applications. Topics for the minisymposium include but are not limited to: 1) deterministic and statistical approaches to inversion problems and data assimilation; 2) reduced order modeling inversion framework; 3) treatment of nonlinear and non-smooth processes; 4) sensitivity analysis and its application to adaptive (targeting) observation; 5) operational data assimilation systems; 6) uncertainties impact studies; 7) automatic tools to support inversion and data assimilation methodologies.
Organizer: Razvan Stefanescu Virginia Tech, USA
Organizer: Adrian Sandu Virginia Tech, USA
Organizer: Ionel M. Navon Florida State University, USA
Organizer: Dacian N. Daescu Portland State University, USA 4:25-4:45 Multi-Scale Data Assimilation for Fine-Resolution Models Zhijin Li, California Institute of Technology, USA 4:50-5:10 Unified EnsembleVariational Data Assimilation System Milija Zupanski, Colorado State University, USA 5:15-5:35 Ensemble Kalman Filters Without Tuning for Large Applications Jeffrey Anderson, National Center for Atmospheric Research, USA 5:40-6:00 Orthogonal Transformations for the Ensemble Kalman Filter Humberto C. Godinez, Earl Lawrence, and Dave Higdon, Los Alamos National Laboratory, USA
2015 SIAM Conference on Computational Science and Engineering
147
Tuesday, March 17
Tuesday, March 17
Tuesday, March 17
MS244
MS245
MS246
Large-Scale and PDEConstrained Optimization 4:25 PM-6:05 PM
Numerical Methods for Multimaterial Flows with Deforming Boundaries Part III of III 4:25 PM-6:05 PM
High-level Technical Computing with Julia 4:25 PM-6:05 PM
Room:251 F We present recent advances in scalable solvers for large-scale and PDE-constrained optimization. Special emphasis is placed on recent developments that extend traditional optimization techniques to matrix-free PDEconstrained case as well as the development of efficient and scalable solvers for PDEconstrained optimization under uncertainty.
Organizer: Sven Leyffer Argonne National Laboratory, USA 4:25-4:45 Inexact Primal-Dual Interior Point Filter Method Victor Zavala, Argonne National Laboratory, USA 4:50-5:10 PDE-Constrained Optimization Under Uncertainty for Convection-Diffusion-Reaction Systems Bart G. Van Bloemen Waanders, Sandia National Laboratories, USA; Harriet Li, Massachusetts Institute of Technology, USA; Drew P. Kouri and Denis Ridzal, Sandia National Laboratories, USA 5:15-5:35 An Adaptive Augmented Lagrangian Method for Large-Scale Constrained Optimization Frank E. Curtis, Lehigh University, USA; Nicholas I.M. Gould, Rutherford Appleton Laboratory, United Kingdom; Hao Jiang and Daniel Robinson, Johns Hopkins University, USA 5:40-6:00 A Data-Driven Approach to PDE-Constrained Optimization Under Uncertainty Drew P. Kouri, Sandia National Laboratories, USA
Room:254 A For Part 2 see MS220 Multimaterial flows with deforming boundaries appear in many applications involving biological flows, medicine, industry, and homeland security. As the number of applications grow, so too the amount of research into developing and improving existing numerical methods for deforming boundary problems in multi-material flows. In this minisymposium, novel numerical methods will be presented that address efficiency and accuracy in adaptive mesh refinement algorithms, scalability on high performance computing platforms, robust interface capturing methods, numerical methods for phase change (freezing, boiling, evaporation), and numerical methods for enforcing boundary conditions at deforming boundary interfaces.
Organizer: Mark Sussman Florida State University, USA
Organizer: Frederic G. Gibou University of California, Santa Barbara, USA
Organizer: Feng Xiao Tokyo Institute of Technology, Japan 4:25-4:45 A Moment-of-Fluid Method for Computing Solutions to Multi-Phase Flows Mark Sussman, Florida State University, USA 4:50-5:10 A Time Splitting Projection Scheme for Compressible Two-Phase Flows: Application to the Interaction of Bubbles and Droplets with Ultrasound Waves Sebastien Tanguy, Universite de Toulouse, France 5:15-5:35 A Finite-Volume Based Formulation for Viscoelastic TwoPhase Flows Shahriar Afkhami, New Jersey Institute of Technology, USA 5:40-6:00 Fourth-Order Interface Tracking and Curvature Estimation for An Arbitrary Number of Materials in Two Dimensions Qinghai Zhang and Aaron L. Fogelson, University of Utah, USA
Room:254 B Dynamic high-level languages enable rapid development of expressive and easily extensible code and therefore are increasingly popular in science and engineering. The new and emerging language Julia aims to overcome some limitations often associated with established dynamic languages, most importantly, large computational overhead and imperfect scaling on parallel computers. This minisymposium covers recent advances and contributions to the Julia language itself as well as examples outlining Julia’s potential for the solution of large-scale real-world problems.
Organizer: Lars Ruthotto Emory University, USA
Organizer: Chen Jiahao Massachusetts Institute of Technology, USA
Organizer: Alan Edelman Massachusetts Institute of Technology and Interactive Supercomputing, Inc., USA 4:25-4:45 Large-Scale 3D Electromagnetic Imaging Using Julia Eldad Haber, University of British Columbia, Canada 4:50-5:10 Distributed and Parallel Computing for Pde Constrained Optimization in Julia Lars Ruthotto, Emory University, USA 5:15-5:35 JuMP: Algebraic Modeling of Optimization Problems in Julia Miles Lubin, Iain Dunning, and Joey Huchette, Massachusetts Institute of Technology, USA 5:40-6:00 An Extensible Test Matrix Collection for Julia Weijian Zhang and Nicholas Higham, University of Manchester, United Kingdom
148
Tuesday, March 17
MS247
2015 SIAM Conference on Computational Science and Engineering
Tuesday, March 17
MS248
Tuesday, March 17
MS249
Student Days: An Informal Meeting 4:25 PM-6:05 PM
Moment Methods for the Boltzmann Equation 4:25 PM-6:05 PM
Low-rank Optimization with Applications - Part II of II 4:25 PM-6:05 PM
Room:255 D
Room:150 AB
Room:150 DE
This informal session provides opportunities for students to meet invited speakers. This is your chance to ask research or career questions, or listen to advice provided by the experts. Discussions will be conducted in smaller groups; after about 20 minutes, the groups will change, so you get a chance to talk to all of the invited speakers present.
The Boltzmann equation is the fundamental model in the kinetic theory. It is useful in a number of high-tech fields such as the aircraft manufacturing and micro-electro-mechanical systems. However, the numerical cost of solving the Boltzmann equation directly in large systems is still unaffordable, and the moment method is one of the important methods to derive simpler models. Recently, the research on the moment method is very active. The purpose of this minisymposium is to gather most representative researchers and report their progress. It invites speakers from different parts of the world and provides a good opportunity to exchange ideas.
For Part 1 see MS224 Low-rank matrices and tensors are a popular modeling tool in machine learning and an effective approximation class in many highdimensional problems. At the same time, low-rank approximations can potentially lead to significant computational savings in algorithms that have to deal with large and dense matrices. This minisymposium showcases recent work in this area by focusing on low-rank optimization. The applications include subspace tracking, nonlinear programming, tensor completion, and high-dimensional eigenvalue problems, while the algorithmic techniques involve greedy algorithms, stochastic optimization, preconditioning, and optimization on manifolds.
Organizer: Hans De Sterck University of Waterloo, Canada
Organizer: Christopher Johnson University of Utah, USA
Organizer: Lois Curfman McInnes Argonne National Laboratory, USA Meet Informally with the CSE15 Co-Chairs and Several Invited Speakers
Organizer: Zhenning Cai RWTH Aachen University, Germany
Organizer: Zhicheng Hu Hong Kong Polytechnic University, China 4:25-4:45 Approximation of the Boltzmann Equation with the Method of Moments for Low Speed Gas Flow Xiaojun Gu, David Emerson, and Jianping Meng, STFC Daresbury Laboratory, United Kingdom 4:50-5:10 A Framework on Moment Model Reduction for Kinetic Equation Ruo Li, Peking University, China 5:15-5:35 Numerical Solution of a Fourteen-Moment Closure for Nonequilibrium Gases James McDonald and Amir Baradaran, University of Ottawa, Canada; Boone Tensuda and Clinton P. Groth, University of Toronto, Canada 5:40-6:00 Theoretical and Computational Investigations of the Non-linear Coupled Constitutive Relations (NCCR) Anirudh Singh Rana and Rho Shin Myong, Gyeongsang National University, Korea
Organizer: Bart Vandereycken Princeton University, USA
Organizer: Daniel Kressner EPFL, Switzerland 4:25-4:45 Exploiting Active Subspaces for Nonlinear Programming Paul Constantine, Colorado School of Mines, USA; David F. Gleich, Purdue University, USA 4:50-5:10 Non-Convex Low-Rank Matrix and Tensor Recovery Yangyang Xu, Rice University, USA 5:15-5:35 Towards an Optimal Scalability in Computing Extreme Eigenpairs of Large Matrices Yin Zhang, Rice University, USA; Zaiwen Wen, Peking University, China; Xin Liu, Chinese Academy of Sciences, China 5:40-6:00 Using Stochastic Optimization Methods for the Polyadic Decomposition of LargeScale Tensors Nico Vervliet and Lieven De Lathauwer, Katholieke Universiteit Leuven, Belgium
2015 SIAM Conference on Computational Science and Engineering
149
Tuesday, March 17
Tuesday, March 17
Tuesday, March 17
MS250
MS251
MS252
Partitioning for Multiple Constraints and Objectives 4:25 PM-6:05 PM
Recent Advances in High Order Spatial Discretization Methods for PDEs Part III of III 4:25 PM-6:05 PM
Efficient Solvers for PDEconstrained Optimization Part II of II 4:25 PM-6:05 PM
Room:260 B For Part 2 see MS228 Many complex physical phenomena can be modeled by partial differential equations and the numerical simulation of PDEs has been an active research area. This minisymposium aims to bring researchers together to discuss the numerical challenges when designing high order spatial discretization methods for time-dependent/static PDEs, with particular attention on weighted essentially non-oscillatory (WENO) methods and discontinuous Galerkin (DG) methods.
For Part 1 see MS229 Optimization problems constrained by partial differential equations arise in a wide range of applications in science and engineering. From a computational viewpoint their solution pose various challenges. One of these challenges is the development of efficient solution methods, which must often be adapted to each specific problem class. The aim of this minisymposium is to bring together researchers to report on recent developments on efficient methods for solving PDE-constrained problems.
Organizer: Yuan Liu
Organizer: Simon W. Funke
Room:258 Traditional graph and hypergraph partitioners play an important role in computational science applications. They enforce a simple balance constraint, with per-part workloads represented by the sum of vertex weights. Edge weights represent communication costs; partitioners then minimize a simple cost objective while satisfying the work constraint. As applications and architectures evolve, the need for partitioners that enforce multiple constraints while trying to satisfy multiple objectives is increasing. This minisymposium showcases recent developments in partitioners that handle multiple constraints and multiple objectives, targetting improved performance for diverse applications such as multiscale mechanics simulations, complex networks, sparse matrix-vector multiplication, and hybrid solvers.
Organizer: Karen D. Devine Sandia National Laboratories, USA
Organizer: Siva Rajamanickam Sandia National Laboratories, USA
Organizer: Kamesh Madduri Pennsylvania State University, USA 4:25-4:45 Load Balancing Multiscale Simulations William R. Tobin, Daniel Fovargue, and Mark S. Shephard, Rensselaer Polytechnic Institute, USA 4:50-5:10 Handling Multiple Communication Metrics for Hypergraph Partitioning Mehmet Deveci and Kamer Kaya, The Ohio State University, USA; Bora Ucar, LIPENS Lyon, France; Umit V. Catalyurek, The Ohio State University, USA 5:15-5:35 Towards a Recursive Graph Bipartitioning Algorithm for Well Balanced Domain Decomposition Astrid Casadei, INRIA Bordeaux Sud-Ouest, France; Pierre Ramet, LABRI, Univ Bordeaux, France; Jean Roman, INRIA, France 5:40-6:00 Complex Objective Partitioning of Small-World Networks Using Label Propagation George Slota and Kamesh Madduri, Pennsylvania State University, USA; Siva Rajamanickam, Sandia National Laboratories, USA
Room:151 AB
Michigan State University, USA
Simula Research Laboratory, Norway
Organizer: Qi Tang
Organizer: Kent-Andre Mardal
Michigan State University, USA
Organizer: Andrew J. Christlieb Michigan State University, USA 4:25-4:45 Maximum Principle and Positivity Preserving Flux Limiters for High Order Schemes Zhengfu Xu, Michigan Technological University, USA 4:50-5:10 A Seventh Order Hybrid Weighted Compact Scheme Based on WENO Stencil for Hyperbolic Conservation Laws Jun Peng and Yiqing Shen, Chinese Academy of Sciences, China 5:15-5:35 High-Order Accurate Numerical Methods for Elliptic and Parabolic Interface Models Yekaterina Epshteyn, University of Utah, USA 5:40-6:00 A One-Stage HighResolution Constrained Transport Method for Magnetohydrodynamic Equations Xiao Feng, Michigan State University, USA
University of Oslo, and Simula Research Laboratory, Norway
Organizer: Marie E. Rognes Simula Research Laboratory, Norway 4:25-4:45 Accelerated SourceEncoding Full-Waveform Seismic Inversion with Additional Constraints Michael Ulbrich and Christian Boehm, Technical University of Munich, Germany 4:50-5:10 PDE-constrained Optimization with Local Control and Boundary Observations: Robust Preconditioners Ole Løseth Elvetun, Norwegian University of Life Sciences, Norway; Bjørn F. Nielsen, Simula Research Laboratory, Norway 5:15-5:35 Robust Preconditioners for PDE-Constrained Optimization with Limited Observation Data Magne Nordaas, Simula Research Laboratory, Norway; Kent-Andre Mardal, University of Oslo, and Simula Research Laboratory, Norway; Bjørn F. Nielsen, Simula Research Laboratory, Norway 5:40-6:00 All-at-once Approach to Optimal Control Problems Constrained by PDEs with Uncertain Inputs Akwum Onwunta, Max Planck Institute, Magdeburg, Germany; Peter Benner, Max Planck Institute for Dynamics of Complex Systems, Germany; Martin Stoll, Max Planck Institute, Magdeburg, Germany
150
Tuesday, March 17
MS253
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Tuesday, March 17
MS254
Turbulence in Wind Turbine Wake: Realistic Turbine/ atmosphere Interaction 4:25 PM-5:40 PM
Advancements in Generalizing Algebraic Multigrid Methods 4:25 PM-6:05 PM
Registration 7:45 AM-2:30 PM
Room:151 DE
Room:151 G
Room:East Foyer
Full-scale horizontal axis wind turbines (WTs) operate in atmospheric boundary layer with atmospheric forcings playing an important role on the wake generated behind the wind turbine. ABL and WT interactions result in strong wake turbulence that adversely impacts the overall performance of wind turbine. Many studies todate have focused on idealistic inflow conditions and have not included the role of atmospheric forcings on the atmosphere on the turbine wake structure. The talks will focus on the effects of atmospheric stratification on the wake structure.
Recent advances in algebraic multigrid methods have led to improved solvers for a wider range of applications. In this collection of talks we highlight some interrelated developments in generalizing interpolation, improving coarsening, and enhancing the setup of multigrid methods. Collectively, techniques such as energy minimization and adaptivity lead to accelerated convergence for challenging applications. In addition, with careful algorithmic decisions and new theoretical observations, these concepts result in a more general framework for algebraic solvers.
Organizer: Kiran Bhaganagar
Organizer: Luke Olson
University of Texas, San Antonio, USA
University of Illinois at Urbana-Champaign, USA
Implications of Numerical and Data Intensive Technology Trends on Scientific Visualization and Analysis 8:15 AM-9:00 AM
Organizer: James Brannick
Room:355
4:25-4:45 Characterizing Turbulence in Wind Turbine Wake: Role of Stratification Kiran Bhaganagar, University of Texas, San Antonio, USA 4:50-5:10 Improved Understanding of Atmospheric Stability Effects on Wind Farm Performance Using Large-Eddy Simulation Cristina L. Archer, Shengbai Xie, and Niranjan Ghaisas, University of Delaware, USA 5:15-5:35 Les Study of a Large Wind Farm Within a Diurnal Atmospheric Boundary Layer Marc Calaf, University of Utah, USA
Pennsylvania State University, USA 4:25-4:45 Root-Node Based Algebraic Multigrid Luke Olson, University of Illinois at UrbanaChampaign, USA; Jacob B. Schroder, Lawrence Livermore National Laboratory, USA 4:50-5:10 Bootstrap and Adaptive Methods James Brannick, Pennsylvania State University, USA 5:15-5:35 Algebraic Multigrid for H-hermitian Matrices Karsten Kahl, University of Wuppertal, Germany 5:40-6:00 Algebraic Multigrid Method for Implicit Smoothed Particle Hydrodynamics Xiaozhe Hu, Tufts University, USA; Wenxiao Pan, Pacific Northwest National Laboratory, USA; Jinchao Xu and Hongxuan Zhang, Pennsylvania State University, USA
Business Meeting 6:15 PM-7:15 PM Room:355 Complimentary wine and beer will be served. (Open to SIAG/CSE members)
Closing Remarks 8:00 AM-8:15 AM Room:355
IP9
Chair: Han-Wei Shen, The Ohio State University, USA Technology trends in numerically and data intensive computing have the potential to reshape and significantly advance how we visualize and analyze the results of scientific simulations. However, next generation numerically intensive supercomputers are bound by power and storage constraints. These require us to transition from standard post-processing visualization and analysis approaches to intelligent, automated in-situ ones. In addition, data intensive technology trends that support accessing and understanding our data using intuitive, webbased and query-driven interfaces are now the norm. In this talk, I will discuss these trends and several freely available, opensource approaches that leverage them.
James Ahrens Los Alamos National Laboratory, USA
Intermission 9:00 AM-9:10 AM
2015 SIAM Conference on Computational Science and Engineering
151
Wednesday, March 18
Wednesday, March 18
Wednesday, March 18
CP1
CP2
CP3
Simulation of Molecular Systems 9:10 AM-10:25 AM
Multigrid Methods 9:10 AM-10:25 AM
Software Tools for Scientific Simulation 9:10 AM-10:25 AM
Room:355
Chair: Matthias Bolten, University of Wuppertal, Germany
Chair: Martin T. Horsch, University of Kaiserslautern, Germany 9:10-9:20 Visualizing Structure in Atomic Systems at High Temperatures Emanuel A. Lazar, Jian Han, and David J. Srolovitz, University of Pennsylvania, USA 9:25-9:35 Potential-Functional Embedding Theory for Seamless Multiscale Simulations of Materials Chen Huang, Florida State University, USA 9:40-9:50 A Numerical and Computational Framework for Hierarchical Multi-Scale/multi-Physics Simulations Jaroslaw Knap, Oleg Borodin, Carrie E. Spear, Kenneth W. Leiter, David A. Powell, and Richard C. Becker, U.S. Army Research Laboratory, USA 9:55-10:05 Amr Strategies for Scft Algorithm Gaddiel Ouaknin and Frederic G. Gibou, University of California, Santa Barbara, USA 10:10-10:20 Computational Molecular Engineering: An Emerging Technology in Process Engineering Martin T. Horsch and Stephan Werth, University of Kaiserslautern, Germany; Christoph Niethammer and Colin Glass, High Performance Computing Center Stuttgart, Germany; Wolfgang Eckhardt, Philipp Neumann, and Hans-Joachim Bungartz, Technische Universität München, Germany; Jadran Vrabec, University of Paderborn, Germany; Hans Hasse, University of Kaiserslautern, Germany
Room:355 A
9:10-9:20 Dependance of the Convergence of Multigrid Methods on the Used Discretization Matthias Bolten, University of Wuppertal, Germany 9:25-9:35 On Teh Efficiency of Nonlinear Multigrid Methods Peter K. Jimack, University of Leeds, United Kingdom 9:40-9:50 Robust Multigrid Methods for Magnetohydrodynamics Thomas Benson, Tufts University, USA 9:55-10:05 Large-Scale Sparse Inverse Covariance Estimation Eran Treister, Javier Turek, and Irad Yavneh, Technion Israel Institute of Technology, Israel 10:10-10:20 Support Graph Smoothing Alyson Fox, University of Colorado Boulder, USA
Room:250 A Chair: Johannes Grotendorst, Forschungszentrum Jülich, Germany 9:10-9:20 Symbolic Representation and Automated Code Generation for Discontinuous Galerkin Finite Element Methods Nathan Sime and Paul Houston, University of Nottingham, United Kingdom 9:25-9:35 Automatic Parallel Programming for Scientific Simulation Li Liao, Aiqing Zhang, and Zeyao Mo, Institute of Applied Physics and Computational Mathematics, China 9:40-9:50 Distributive Interoperable Executive Library (diel) for Systems of Multiphysics Simulation Kwai L. Wong, University of Tennessee and Oak Ridge National Laboratory, USA; Jason Coan and David White, Maryville College, USA 9:55-10:05 PoKiTT: an Efficient, Platform Agnostic Package for Thermodynamics, Kinetics, and Transport Properties within PDE Solvers Nathan Yonkee and James C. Sutherland, University of Utah, USA 10:10-10:20 CSE Education at JSC Johannes Grotendorst, Forschungszentrum Jülich, Germany
152
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
Wednesday, March 18
CP4
CP5
CP6
Parallel Dense Linear Algebra 9:10 AM-10:10 AM
Bayesian Methods in Uncertainty Quantification 9:10 AM-10:25 AM
Preconditioners for Fluids, Structures and Images 9:10 AM-10:25 AM
Room:250 B
Room:250 C
Room:250 D
Chair: Eduardo F. D’Azevedo, Oak Ridge National Laboratory, USA
Chair: Isabell Franck, Technical University of Munich, Germany
Chair: Kees Vuik, Delft University of Technology, Netherlands
9:10-9:20 Dynamic Runtime Scheduling for Dense Out-of-Core Matrix Computation on the Intel Xeon Phi Eduardo F. D’Azevedo, Oak Ridge National Laboratory, USA; Ben Chan and Terrence Chong, Chinese University of Hong Kong, Hong Kong; Allan Morales, George Washington University, USA; Kwai L. Wong, University of Tennessee and Oak Ridge National Laboratory, USA
9:10-9:20 Uncertainty Propagation Using Infinite Mixture of Gaussian Processes and Variational Bayesian Inference Peng Chen, Cornell University, USA; Nicholas Zabaras, University of Warwick, United Kingdom; Ilias Bilionis, Purdue University, USA
9:10-9:20 Indefinite Preconditioning of the Coupled Stokes-Darcy System Scott Ladenheim, Temple University, USA; Prince Chidyagwai, Loyola University, USA; Daniel B. Szyld, Temple University, USA
9:25-9:35 A Framework for Parallel Fast Matrix Multiplication Austin Benson, Stanford University, USA; Grey Ballard, Sandia National Laboratories, USA 9:40-9:50 Optimization of Singular Vectors Computation Sergey V Kuznetsov, Intel Corporation, Russia; Nadezhda Mozartova, Intel Corporation, USA 9:55-10:05 Performance Study of a Randomized Dense Low-Rank Matrix Approximation Using Multiple Gpus Theo Mary, Universite de Toulouse, France; Ichitaro Yamazaki, Jakub Kurzak, Piotr Luszczek, Stanimire Tomov, and Jack J. Dongarra, University of Tennessee, Knoxville, USA
9:25-9:35 Minimal Set of Mechanisms Controlling Type I Interferon Differential Signaling Pencho Yordanov, Irene Otero-Muras, and Joerg Stelling, ETH Zürich, Switzerland 9:40-9:50 Probabilistic Mixtures of Local Reduced-Basis Models for Uncertainty Quantification Michael Kraus and Phaedon S. Koutsourelakis, Technical University of Munich, Germany 9:55-10:05 Surrogate-Based Bayesian Model Ranking of Atomistic Models Incorporating the Fidelity of Surrogates Hadi Meidani, University of Illinois at Urbana-Champaign, USA; Mike Kirby and Dmitry Bedrov, University of Utah, USA 10:10-10:20 Variational Bayesian Formulations for High-Dimensional Inverse Problems Isabell Franck and Phaedon S. Koutsourelakis, Technical University of Munich, Germany
9:25-9:35 A Parallel Linear Solver Exploiting the Physical Properties of the Underlying Mechanical Problem Kees Vuik, Delft University of Technology, Netherlands 9:40-9:50 Preconditioner Scaling for Finite Element Models of Turbulent Air/Water Flow in Coastal and Hydraulic Applications Chris Kees and Aron Ahmadia, U.S. Army Engineer Research and Development Center, USA; Jed Brown, Argonne National Laboratory, USA and University of Colorado Boulder, USA; Matthew Farthing, U.S. Army Engineer Research and Development Center, USA; Barry F. Smith, Argonne National Laboratory, USA 9:55-10:05 A Scalable NewtonKrylov-Schwarz Method for Coupled Fluid-Structure Interaction Problems Fande Kong and Xiao-Chuan Cai, University of Colorado Boulder, USA 10:10-10:20 Data Based Regularization Methods Thomas K. Huckle, Technische Universität München, Germany
2015 SIAM Conference on Computational Science and Engineering
153
Wednesday, March 18
Wednesday, March 18
Wednesday, March 18
CP7
CP8
CP9
Numerical Simulation on Accelerators 9:10 AM-10:25 AM
Parallel Iterative Methods with Reduced Communication 9:10 AM-10:25 AM
Numerical PDEs I 9:10 AM-10:25 AM
Room:250 E Chair: Erik G. Boman, Sandia National Laboratories, USA 9:10-9:20 Spectral Methods in PDE Solving: a Multi-GPU Framework Gyula I. Toth, Tatjana Kuztensova, and Bjorn Kvamme, University of Bergen, Norway 9:25-9:35 Direct Hierarchical Schur Method for Nested Dissection Reordered Linear Systems on MultiGPUs Cheming Chu, Pochuan Wang, and Weichung Wang, National Taiwan University, Taiwan 9:40-9:50 Solving Sparse Linear Systems on GPUs Based on the Biell Storage Format Tongxiang Gu, Institute of Applied Physics and Computational Mathematics, China; Cong Zheng, Graduate School of Chinese Academy of Engineering Physics, China; Shou Gu, University of Electronic Science and Technology of China, China; Xingping Liu, Institute of Applied Physics and Computational Mathematics, China 9:55-10:05 Optimizing Structured Grid Numerical Simulations for NumaMulticore Systems Zhang Yang, Aiqing Zhang, and Zeyao Mo, Institute of Applied Physics and Computational Mathematics, China 10:10-10:20 Parallel Graph Coloring for Scientific Computing Erik G. Boman, Sandia National Laboratories, USA
Room:250 F Chair: Grey Ballard, Sandia National Laboratories, USA 9:10-9:20 Multigrid Preconditioners for Communication-Avoiding Krylov Methods Andrey Prokopenko, Sandia National Laboratories, USA 9:25-9:35 Reducing Communication Costs for Sparse Matrix Multiplication within Algebraic Multigrid Grey Ballard, Jonathan J. Hu, and Christopher Siefert, Sandia National Laboratories, USA
Room:251 A Chair: Anna Lischke, Iowa State University, USA 9:10-9:20 Asymptotic-Preserving Space-Time Discontinuous Galerkin Schemes for a Class of Relaxation Systems Anna Lischke and James A. Rossmanith, Iowa State University, USA 9:25-9:35 Universal Meshes for Problems with Moving Boundaries Evan S. Gawlik and Adrian Lew, Stanford University, USA
9:40-9:50 Alpha Setup-Amg: An Adaptive Setup Based Amg Solver for Large-Scale Simulations with Long Time-Stepping Xiaowen Xu, Institute of Applied Physics and Computational Mathematics, China
9:40-9:50 A Multiscale Finite Volume with Oversampling Method to Simulate Low-Frequency Electromagnetic Geophysical Responses Luz Angelica A. Caudillo Mata and Eldad Haber, University of British Columbia, Canada; Lars Ruthotto, Emory University, USA; Christoph Schwarzbach, University of British Columbia, Canada
9:55-10:05 Avoiding Communication and Synchronization in Krylov Eigensolvers Alexander Breuer, Claire Eisner, Jaroslaw Knap, and Kenneth Leiter, U.S. Army Research Laboratory, USA
9:55-10:05 A Computational ShockTube for Reproducible Computational Experiments in Traumatic Brain Injury Mauricio J. Del Razo and Randall LeVeque, University of Washington, USA; David Cook, VA Hospital, USA
10:10-10:20 Scalable Alternative to Domain Decomposition David A. Appelhans, Thomas Manteuffel, Steve McCormick, and John Ruge, University of Colorado Boulder, USA
10:10-10:20 Numerical Simulations of Biological Invasions Shilpa Khatri, University of California, Merced, USA; Anna-Karin Tornberg, KTH Royal Institute of Technology, Sweden
154
Wednesday, March 18
CP10 Numerical PDEs II 9:10 AM-10:25 AM Room:251 B Chair: Saeid Karimi, University of Houston, USA 9:10-9:20 Towards a Physically Admissible Implicit-Explicit Splitting for All Froude Number Shallow Water Flows Seyed Hamed Zakerzadeh and Sebastian Noelle, RWTH Aachen University, Germany 9:25-9:35 High Order Schemes Based on Operator Splitting and Deferred Corrections for Stiff Time Dependent Pdes Max Duarte and Matthew Emmett, Lawrence Berkeley National Laboratory, USA 9:40-9:50 Monolithic Multi-TimeStep Coupling Methods for Transient Systems Saeid Karimi and Kalyana Nakshatrala, University of Houston, USA 9:55-10:05 A New Lattice Boltzman Solver on Unstructured Grid and Study of Its Performance Weishan Deng, Xiaohe Zhufu, and Jin Xu, Chinese Academy of Sciences, China 10:10-10:20 Finite-Difference Frequency-Domain Analysis of Photonic Devices with Periodic Structures Based on Domain Decomposition Cheng-Han Du, Pochuan Wang, and Weichung Wang, National Taiwan University, Taiwan
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
CP11
CP12
Parallel Simulation with Reduced Communication 9:10 AM-10:25 AM
Numerical Methods for Statistical Mechanics and Plasmas 9:10 AM-10:25 AM
Room:251 C Chair: Miles L. Detrixhe, University of California, Santa Barbara, USA
Room:251 D
9:10-9:20 Exploring Communication Options with Adaptive Mesh Refinement Courtenay T. Vaughan and Richard Barrett, Sandia National Laboratories, USA
9:10-9:20 Vlasov-Poisson Simulations of Magnetized Plasmas Using HighOrder Continuum Methods Genia Vogman, University of California, Berkeley, USA; Phillip Colella, Lawrence Berkeley National Laboratory, USA; Uri Shumlak, University of Washington, USA
9:25-9:35 A Communication Algorithm for the Patch-Based Multiblock Structured Mesh Applications Hong Guo, Institute of Applied Physics and Computational Mathematics, China 9:40-9:50 A Communication Staging Technique for a Hierarchical Ocean Model Geoff Womeldorff, Chris Newman, Dana Knoll, and Luis Chacón, Los Alamos National Laboratory, USA 9:55-10:05 Scalable Parallel Assembly for High-Performance Computing with Isogeometric and Higher-Order Finite Elements Vasco Varduhn and Dominik Schillinger, University of Minnesota, USA 10:10-10:20 A Parallel Fast Sweeping Method for Quadtrees and Octrees Miles L. Detrixhe, University of California, Santa Barbara, USA
Chair: Paul Cazeaux, EPFL, Switzerland
9:25-9:35 Coarse Multiscale Timestepping for Problems in Plasma Physics with Equation-Free Projective Integration Paul Cazeaux, EPFL, Switzerland; Jan Hesthaven, EPFL, France 9:40-9:50 Discontinuous Galerkin Deterministic Solvers of BoltzmannPoisson Models of Hot Electronic Transport Using Empirical Pseudopotential Methods Jose A. Morales Escalante and Irene M. Gamba, University of Texas at Austin, USA; Yingda Cheng, Michigan State University, USA; Armando Majorana, University of Catania, Italy; Chi-Wang Shu, Brown University, USA; James R. Chelikowsky, University of Texas at Austin, USA 9:55-10:05 Semi-Lagrangian Discontinuous Galerkin Schemes for the Relativistic Vlasov-Maxwell System Pierson Guthrey and James A. Rossmanith, Iowa State University, USA 10:10-10:20 Asymptotic-Preserving Scheme for the Fokker-PlanckLandau-Maxwell System in the QuasiNeutral Regime Stephane Brull, Institut Polytechnique de Bordeaux, France; Bruno Dubroca, d’Humière Emmanuel, and Guisset Sébastien, Universite de Bordeaux I, France
2015 SIAM Conference on Computational Science and Engineering
155
Wednesday, March 18
Wednesday, March 18
Wednesday, March 18
CP13
CP14
CP15
Uncertainty Quantification 9:10 AM-10:25 AM
Numerical Methods for Control Problems 9:10 AM-10:25 AM
Applications to Energy Systems and Signals 9:10 AM-9:40 AM
Chair: Uno B. Vaaland, Norwegian University of Science and Technology, Norway
Room:251 F
Room:254 A
Chair: Ana Maria Soane, US Naval Academy, USA
Chair: Duncan A. Mcgregor, Oregon State University, USA
9:10-9:20 A Multi-Model Monte Carlo Framework Based on Ensemble Kalman Filtering Jianxun Wang and Heng Xiao, Virginia Tech, USA
9:10-9:20 Optimal Order Multigrid Preconditioners for Linear Systems Arising in the Semismooth Newton Method Solution Process of a Class of Control-Constrained Problems Jyoti Saraswat, Thomas More College, USA; Andrei Draganescu, University of Maryland, Baltimore County, USA
9:10-9:20 Simulation-based Current Estimation in Magnetohydrodynamic Generators Duncan A. Mcgregor, Vrushali Bokil, and Nathan L. Gibson, Oregon State University, USA; Charles Woodside, National Energy Technology Laboratory, USA
9:25-9:35 Multigrid Preconditioners for Stochastic Optimal Control Problems with Elliptic Spde Constraints Ana Maria Soane, US Naval Academy, USA
9:25-9:35 Fast Supercomputing Algorithms for Power System Operation and Control Eugene A. Feinberg, Stony Brook University, USA; Bruce Fardanesh, New York Power Authority, USA; Muqi Li and Roman Samulyak, Stony Brook University, USA; George Stefopoulos, New York Power Authority, USA; Gaurish Telang, Stony Brook University, USA
Room:251 E
9:25-9:35 Detecting Discontinuities and Localized Features Using Gaussian Processes Ilias Bilionis, Purdue University, USA; Nicholas Zabaras, Cornell University, USA 9:40-9:50 Reducing Dimensionality Through Active Subspaces, and the Effect of Gradient Approximations on the Associated Eigenpairs Uno B. Vaaland, Norwegian University of Science and Technology, Norway; Paul Constantine, Colorado School of Mines, USA 9:55-10:05 Parallel Methods for Accelerated Multilevel Monte Carlo for Partial Differential Equations with Random Input Zane Colgin, Middle Tennessee State University, USA 10:10-10:20 Topology Optimization under Manufacturing Uncertainties Boyan S. Lazarov, Technical University of Denmark, Denmark
9:40-9:50 Optimal Control of Level Sets Christopher Basting and Dmitri Kuzmin, Technische Universität Dortmund, Germany 9:55-10:05 Numerical Realization of the Open Pit Mine Planning Problem Nikolai Strogies and Andreas Griewank, Humboldt University Berlin, Germany 10:10-10:20 Fractional Powers of Finite Element Approximation for An Parabolic Optimal Control Problems Manickam Kandasamy and Periasamy Prakash, Periyar University, India
156
Wednesday, March 18
CP16 Applications in Geophysics 9:10 AM-10:25 AM Room:254 B Chair: Fabian Franzelin, University of Stuttgart, Germany 9:10-9:20 Data-Driven Uncertainty Quantification with Adaptive Sparse Grids in Subsurface Flow Simulations Fabian Franzelin and Sergey Oldayshkin, University of Stuttgart, Germany; Benjamin Peherstorfer, Massachusetts Institute of Technology, USA; Dirk Pflüger, Universität Stuttgart, Germany 9:25-9:35 Quantification of Structural Uncertainty in a Land Surface Model Zhangshuan Hou and Maoyi Huang, Pacific Northwest National Laboratory, USA; Jaideep Ray and Laura Swiler, Sandia National Laboratories, USA 9:40-9:50 HPC and Model Reduction Algorithms for Large-Scale Simulation of Stochastic Wave Propagation Models Mahadevan Ganesh, Colorado School of Mines, USA 9:55-10:05 Integration of Geophysical Fluid Dynamics and Fully 3D Fluid Dynamics to Simulate Multiphysics Coastal Ocean Flows Hansong Tang, City University of New York, USA; Ke Qu, City College of New York, USA 10:10-10:20 PDE-Constrained Optimization Applied to Core Flooding from Reservoir Engineering Caleb C. Magruder, Rice University, USA; Jeremy Brandman and Shivakumar Kameswaran, ExxonMobil Research, USA
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
CP17
CP18
Incompressible and Low Mach-number Fluids 9:10 AM-10:25 AM
FEM for Fluids and Structures 9:10 AM-10:25 AM
Room:254 C
Chair: Kathrin Smetana, Massachusetts Institute of Technology, USA
Chair: Suchuan Dong, Purdue University, USA
Room:150 AB
9:10-9:20 Energy-Stable Open Boundary Conditions for Two-Phase Flows Suchuan Dong, Purdue University, USA
9:10-9:20 Dual-Mixed Finite Element Methods for the Brinkman Problem Jason Howell, College of Charleston, USA; Noel J. Walkington, Carnegie Mellon University, USA
9:25-9:35 Improving the FluidStructure Interaction for Ship Hydrodynamics Thomas Miras, Fernando A. Rochinha, Renato N. Elias, José L.D. Alves, Carlos E. Silva, and Alvaro L.G.A. Coutinho, Federal University of Rio de Janerio, Brazil
9:25-9:35 Numerical Modeling of Non-Associated Flow Model by Successive Convex Optimization: Application in Incompressible Porous Media Zahra S. Lotfian and Mettupalayam Sivaselvan, State University of New York at Buffalo, USA
9:40-9:50 A Stable Projection Method for the Incompressible Navier-Stokes Equations on Arbitrary Geometries and Adaptive Quad/oc-Trees Arthur Guittet, University of California, Santa Barbara, USA
9:40-9:50 High-Order Mixed Finite Elements for a Pressure Poisson Equation Reformulation of the Navier-Stokes Equations with Electric Boundary Conditions Dong Zhou, Temple University, USA; David Shirokoff, New Jersey Institute of Technology, USA; Benjamin Seibold, Temple University, USA; Rodolfo R. Rosales, Massachusetts Institute of Technology, USA; Prince Chidyagwai, Loyola University, USA
9:55-10:05 A Low Mach Number Model for Moist Atmospheric Flows Max Duarte, Ann S. Almgren, and John B. Bell, Lawrence Berkeley National Laboratory, USA 10:10-10:20 An Efficient, Pressure Projection Method for Reacting LowMach Flow Simulations Amir Biglari, Tony Saad, and James C. Sutherland, University of Utah, USA
9:55-10:05 A Variational Multi-Scale Approach Using Linear Simplicial Finite Elements for Transient Viscoelastic Solid Mechanics Xianyi Zeng and Guglielmo Scovazzi, Duke University, USA 10:10-10:20 Physically Motivated and Certified Approximation of Large Elastic Structures in Real-Time Kathrin Smetana, Massachusetts Institute of Technology, USA; Phuong Huynh and David Knezevic, Akselos, Switzerland; Anthony T. Patera, Massachusetts Institute of Technology, USA
2015 SIAM Conference on Computational Science and Engineering
157
Wednesday, March 18
Wednesday, March 18
Wednesday, March 18
CP19
CP20
CP21
Finite Element Methods 9:10 AM-10:25 AM
Modeling at Molecular Scales in Physics and Chemistry 9:10 AM-10:10 AM
Numerical Models for Fluids 9:10 AM-10:25 AM
Room:258
9:10-9:20 Brownian and Hydrodynamic Motion of Complex Shaped Particles in Straight and Branching Blood Vessels Yaohong Wang, David Eckmann, Ravi Radhakrish, Helena Vitoshki, and Portonovo Ayyaswamy, University of Pennsylvania, USA
Room:150 CD Chair: Emily Evans, Brigham Young University, USA 9:10-9:20 The Weighted Finite Element Method for Elasticity Problem with Singularity Viktor Rukavishnikov, Russian Academy of Sciences, Russia 9:25-9:35 An Anchored Analysis of Variance Petrov-Galerkin Projection Scheme for a Class of High Dimensional Elliptic Partial Differential Equations Matthew T. Li, Christophe Audouze, and Prasanth B. Nair, University of Toronto, Canada 9:40-9:50 Hierarchical Hpk-Adaptivity for Isogeometric Analysis Emily Evans and Kevin Tew, Brigham Young University, USA 9:55-10:05 A Stencil Based Finite Element Method Johan S. Hysing, Tokyo Institute of Technology, Japan 10:10-10:20 Finite Element Modeling and Analysis of Invisibility Cloaks with Metamaterials Jichun Li, University of Nevada, Las Vegas, USA
Chair: Yongyong Cai, Purdue University, USA 9:10-9:20 Ground States and Dynamics of Spin-Orbit-Coupled Bose-Einstein Condensates Yongyong Cai, Purdue University, USA; Weizhu Bao, National University of Singapore, Singapore 9:25-9:35 Fractional Schrödinger Dynamics Yanzhi Zhang, Missouri University of Science and Technology, USA 9:40-9:50 Fast Ewald Summation for Mixed Periodic Boundary Conditions Based on the Nonuniform Fft Franziska Nestler and Michael Pippig, Chemnitz University of Technology, Germany 9:55-10:05 Parallel Replica Dynamics with Spatial Parallelization for a Driven System Michael T. Stobb, Juan Meza, and Ashlie Martini, University of California, Merced, USA
Room:259 Chair: TBA
9:25-9:35 An ALE-Phase-Field Method for Dynamic Wetting of Moving Particles Pengtao Yue, Virginia Tech, USA 9:40-9:50 Interaction Between Toroidal Swimmers in Stokes Flow Jianjun Huang, Worcester Polytechnic Institute, USA; Lisa J. Fauci, Tulane University, USA 9:55-10:05 Simulating Non-Dilute Transport in Porous Media Using a Tcat-Based Model Deena H. Giffen, North Carolina State University, USA 10:10-10:20 Hierarchical Model Reduction of the Navier-Stokes Equations for Incompressible Flows in Pipes Sofia Guzzetti, Emory University, USA; Simona Perotto, Politecnico di Milano, Italy; Alessandro Veneziani, Emory University, USA
158
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
CP23
CP24
Stochastic Differential Equations 9:10 AM-10:25 AM
Numerical Methods for Conservation Laws 9:10 AM-10:25 AM
Models for Geophysical Fluids 9:10 AM-10:25 AM
Room:260 A
Room:260 B
Room:151 AB
Chair: Christophe Audouze, University of Toronto, Canada
Chair: Rakesh Kumar, Indian Institute of Technology-Bombay, India
Chair: Markus Burkow, University of Bonn, Germany
9:10-9:20 Local Polynomial Chaos Expansion for Linear Differential Equations with High Dimensional Random Inputs Yi Chen, Purdue University, USA; Dongbin Xiu, University of Utah, USA; John D. Jakeman, Sandia National Laboratories, USA; Claude Gittelson, ETH Zürich, Switzerland
9:10-9:20 A Space-Time Finite Volume Differencing Method for Robust Higher Order Schemes for Transport Equations Yaw Kyei, North Carolina Agricultural and Technical State University, USA
9:10-9:20 A Numerical Simulation of the Sediment Dynamics in a Three Dimensional Fluid Flow Markus Burkow and Michael Griebel, University of Bonn, Germany
Wednesday, March 18
CP22
9:25-9:35 Stochastic Low-Dimensional Modeling of Natural Convection Using Dynamically Orthogonal Decomposition Hessameddin Babaee, Massachusetts Institute of Technology, USA 9:40-9:50 Anchored ANOVA PetrovGalerkin (AAPG) Projection Schemes for Parabolic Stochastic Partial Differential Equations Christophe Audouze and Prasanth B. Nair, University of Toronto, Canada 9:55-10:05 Fully Implicit Runge-Kutta Methods for Multi-Channel Stiff Stochastic Differential Systems with Jumps Viktor Reshniak and Abdul Khaliq, Middle Tennessee State University, USA; Guannan Zhang, Oak Ridge National Laboratory, USA; David A. Voss, Western Illinois University, USA 10:10-10:20 Variance Reduction in the Simulation of Stochastic Differential Equations David J. Horntrop, New Jersey Institute of Technology, USA
9:25-9:35 Cubic B-Spline QuasiInterpolation Based Numerical Scheme for Hyperbolic Conservation Laws Rakesh Kumar and Sambandam Baskar, Indian Institute of Technology-Bombay, India
9:25-9:35 Three-Dimensional WaveletBased Adaptive Mesh Refinement Algorithm for Numerical Simulation of Atmospheric Global Chemical Transport Artem N. Semakin and Yevgenii Rastigejev, North Carolina A&T State University, USA
9:40-9:50 Lagrangian Particle Method for Complex Flows Roman Samulyak, Hsin-Chiang Chen, and Wei Li, Stony Brook University, USA
9:40-9:50 Free Surface Waves on a Horizontal Shear Flow John P. Mchugh, University of New Hampshire, USA; Gary Lapham, Maine Maritime Academy, USA
9:55-10:05 A Runge-Kutta Discontinuous Galerkin Method for Modeling Storm-Water Flow in Networks of Drainage Channels Prapti Neupane and Clint Dawson, University of Texas at Austin, USA
9:55-10:05 Openfoam Implementation of a New Subgrid-Scale Model for Large Eddy Simulation Rukiye Kara, Mimar Sinan Fine Arts University, Turkey; Mine Caglar, Koc University, Turkey
10:10-10:20 Space-Time Adaptive Multiresolution Simulations of the Compressible Euler Equations Margarete O. Domingues, Instituto Nacional de Pesquisas Espaciais, Brazil
10:10-10:20 Quantifying Scale Coupling and Energy Pathways in the Ocean Hussein Aluie, University of Rochester, USA; Matthew Hecht, Los Alamos National Laboratory, USA; Geoffrey Vallis, University of Exeter, United Kingdom
2015 SIAM Conference on Computational Science and Engineering
159
Wednesday, March 18
Wednesday, March 18
Wednesday, March 18
CP25
CP26
MS255
Linear Algebra Methods 9:10 AM-10:25 AM
Applications in Computational Science 9:10 AM-10:10 AM
Featured Minisymposium: Big Data Analytics 10:55 AM-12:35 PM
Room:150 G
Room:355
Chair: William F. Mitchell, National Institute of Standards and Technology, USA
Big data analytics has become a new paradigm for decision making and knowledge discovery. This minisymposium highlights four major topics, urban, graph, biomedical, and scientific data analytics in this area. Urban data analytics harvest the vast amount of data collected from smart devices and networks to improve the life of citizens. Graphs are used to model relationships between entities and hence play a central role in big data analytics. Biomedical data analytics use data-driven approaches to enable discovery and find new medical cures. Scientific data analytics assists scientists to extract knowledge from simulations aiming to solve grand challenge problems.
Room:151 DE Chair: Daniel B. Szyld, Temple University, USA 9:10-9:20 Efficient Low-Rank Solutions of Generalized Lyapunov Equations Daniel B. Szyld, Temple University, USA; Setephen Shank, Massachusetts Institute of Technology, USA; Valeria Simoncini, Universita’ di Bologna, Italy 9:25-9:35 An Implementation and Analysis of the Refined Projection Method For (Jacobi-)Davidson Type Methods Lingfei Wu and Andreas Stathopoulos, College of William & Mary, USA 9:40-9:50 On a priori and a posteriori Eigenvalue/eigenvector Error Estimates for Nonlinear Eigenvalue Problems Agnieszka Miedlar, Technische Universität Berlin, Germany; Daniel Kressner, EPFL, Switzerland 9:55-10:05 Lu and Partial Orthogonalization Preconditioning for Conjugate Gradient Solution of Overdetermined Sparse Least Squares Problems Gary W. Howell, North Carolina State University, USA; Marc Baboulin, INRIA/ University of Paris-Sud, France 10:10-10:20 Tensor Rank Prediction via Cross Validation Woody N. Austin, University of Texas at Austin, USA; Tamara G. Kolda and Todd Plantenga, Sandia National Laboratories, USA
9:10-9:20 Updating and Downdating Techniques for Networks Francesca Arrigo, Università degli Studi dell’Insubria, Italy; Michele Benzi, Emory University, USA 9:25-9:35 Using Space Filling Curves to Find An Element That Contains a Given Point William F. Mitchell, National Institute of Standards and Technology, USA 9:40-9:50 Dynamic Causal Modelling of Brain-Behaviour Relationships Jean Daunizeau and Lionel Rigoux, Université Pierre et Marie Curie, France 9:55-10:05 Topology Backs Holistic Medicine Fernando Schwartz, University of Tennessee, USA; Louis Xiang, Chinese University of Hong Kong, Hong Kong; Kwai L. Wong, University of Tennessee and Oak Ridge National Laboratory, USA
Coffee Break 10:25 AM-10:55 AM Room:355
Organizer: Han-Wei Shen The Ohio State University, USA 10:55-11:15 Exploring Big Urban Data Claudio T. Silva, New York University, USA 11:20-11:40 Designing Visualizations for Biological Research Miriah Meyer, University of Utah, USA 11:45-12:05 Graph Analytics for Scientific Data Peterka Tom, Argonne National Laboratory, USA 12:10-12:30 Exascale Scientific Data Analytics and Visualization Han-Wei Shen, The Ohio State University, USA
160
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS257
MS258
Distributed Cyber-Physical Systems: Modelling and Controlling the Power Grid Part I of II 10:55 AM-12:35 PM
Recent Advances in Numerical Methods for Interface Problems Part III of IV 10:55 AM-12:35 PM
Streamlining Application Performance Portability Part I of II 10:55 AM-12:35 PM
Room:355 A
Room:250 A
For Part 2 see MS281 The Power Grid is a geographically distributed cyber-physical system, and its system complexity and the volume of data associated with it are growing. The speakers in the minisymposium will address recent advances in algorithms that accelerate the performance of software tools developed to model, optimize, and control the Grid.
For Part 2 see MS233 For Part 4 see MS282 Many real world applications are characterized by multiple materials and complex fluids, and they often lead to interface problems. It is imperative to develop efficient and stable numerical methods for these problems. Great efforts have been made for solving interface problems and tracing the moving interfaces in the past decades. However, many challenges, such as the lack of regularity of physical solutions due to the discontinuity across interfaces, still hinder the development of efficient numerical methods. This minisymposium intends to create a forum for researchers from different fields to discuss recent advances on the interface methods and their applications.
For Part 2 see MS283 Achieving sustainable performance among different computer architectures often cannot be addressed with ingenuous combinations of compiler optimization options and manual rewriting. In contrast, auto-tuning (AT) technologies based, for example, on static modifications (typically pragmas), runtime optimizations, and source to source transformations have enabled sustained performance on a variety of computer architectures. As we move towards more complex computer architectures and higher levels of concurrency, AT is expected to play a major role in giving applications a gateway to performance portability. Of particular interest is the use of AT in the context of kernels that are the core of most large applications, including linear solvers, eigensolvers and stencil computations. In this minisymposium we will discuss AT frameworks and technologies, show examples of how they have been used in practice, and the corresponding performance gains.
Wednesday, March 18
MS256
Organizer: Alex Pothen Purdue University, USA
Organizer: Vaithianathan Venkatasubramanian Washington State University, USA
Organizer: Mahantesh Halappanavar Pacific Northwest National Laboratory, USA 10:55-11:15 Fast Algorithms for Synchrophasor Computations Vaithianathan Venkatasubramanian, Tianying Wu, Seyed Arash Sarmadi, and Ebrahim Rezaei, Washington State University, USA 11:20-11:40 Exploiting Network Laplacian Structure in Power Grid Dynamics Christopher DeMarco and Honghao Zheng, University of Wisconsin, Madison, USA 11:45-12:05 Distributed Optimization Algorithms for Wide-Area Oscillation Monitoring in Power Systems Aranya Chakrabortty and Seyedbehzad Nabavi, North Carolina State University, USA 12:10-12:30 Probability Density Methods for the Analysis of Power Grids Under Uncertainty David A. Barajas-Solano and Alexander Tartakovsky, Pacific Northwest National Laboratory, USA; Debojyoti Ghosh, Emil M. Constantinescu, and Shrirang Abhyankar, Argonne National Laboratory, USA; Zhenyu Huang, Pacific Northwest National Laboratory, USA
Organizer: Xiaoming He Missouri University of Science and Technology, USA
Organizer: Xiaolin Li State University of New York, Stony Brook, USA 10:55-11:15 Optimal Energy Conserving Discontinuous Galerkin Methods for the Wave Propagation Problems in Heterogeneous Media Yulong Xing, University of Tennessee and Oak Ridge National Laboratory, USA 11:20-11:40 Surface Phase Separation Mediated by Nonlocal Interactions Yongcheng Zhou, Colorado State University, USA 11:45-12:05 Improvements in the Level Set Method with a Focus on Curvature-Dependent Forcing Chris Vogl, University of Washington, USA
Room:250 B
Organizer: Takahiro Katagiri University of Tokyo, Japan
Organizer: Toshiyuki Imamura RIKEN, Japan
Organizer: Osni A. Marques Lawrence Berkeley National Laboratory, USA 10:55-11:15 Towards Auto-tuning in the Era of 200+ Thread Parallelisms --- FIBER Framework and Minimizing Software Stack --Takahiro Katagiri, Satoshi Ohshima, and Masaharu Matsumoto, University of Tokyo, Japan
12:10-12:30 A Multi-physics Domain Decomposition Method for NavierStokes-Darcy Model Xiaoming He, Missouri University of Science and Technology, USA
continued on next page
2015 SIAM Conference on Computational Science and Engineering
11:20-11:40 Active Harmony: Making Autotuning Easy Jeffery Hollingsworth, University of Maryland, USA 11:45-12:05 The Role of Autotuning Compiler Technology Mary Hall, University of Utah, USA 12:10-12:30 A Framework for Separation of Concerns Between Application Requirements and System Requirements Hiroyuki Takizawa, Shoichi Hirasawa, and Hiroaki Kobayashi, Tohoku University, Japan
161
Wednesday, March 18
Wednesday, March 18
MS259
MS260
Parallel, Multiscale, Multiphysics Simulation Using MOOSE - Part I of II 10:55 AM-12:35 PM
Recent Advances in Model Reduction - Part IV of V 10:55 AM-12:35 PM
Room:250 C
For Part 3 see MS236 For Part 5 see MS285 Model reduction has become an increasingly important tool to mitigate the computational burden of modeling and simulation in timecritical (e.g., model-predictive control) and many query (e.g., Bayesian inversion) applications. This minisymposium presents recent advances that address the primary challenges facing such methods, such as preserving intrinsic problem structure, handling high- dimensional parameter spaces, integration with existing simulation codes, stability, optimization and uncertainty quantification and control.
For Part 2 see MS284 The Idaho National Laboratory (INL)developed Multiphysics Object Oriented Simulation Environment (MOOSE; www. mooseframework.org), is an open-source, parallel computational platform that enables the solution of complex, fully-implicit multiphysics systems. MOOSE provides a set of computational tools that scientists and engineers can use to create sophisticated multiphysics simulations. Applications built using MOOSE have computed solutions for chemical reaction and transport equations, computational fluid dynamics, solid mechanics, heat conduction, mesoscale materials modeling, geomechanics, and others. This minisymposium highlights some of the recent results obtained using the MOOSE platform.
Organizer: Derek R. Gaston Idaho National Laboratory, USA
Organizer: Andrew Slaughter Idaho National Laboratory, USA 10:55-11:15 Computational Microstructure Science Using the Moose Framework Bradley Fromm, Daniel Schwen, and Michael Tonks, Idaho National Laboratory, USA 11:20-11:40 Microstructural Modeling of Fracture in Uranium Dioxide Using a Phase-Field Based Approach Pritam Chakraborty and Michael Tonks, Idaho National Laboratory, USA 11:45-12:05 Fission Bubble Modeling in Uranium Carbide Christopher Matthews and Andrew Klein, Oregon State University, USA 12:10-12:30 Grizzly: A Simulation Tool for Nuclear Power Plant Component Aging Benjamin Spencer, Idaho National Laboratory, USA
Room:250 D
Organizer: Kevin T. Carlberg Sandia National Laboratories, USA
Organizer: Gianluigi Rozza SISSA, International School for Advanced Studies, Trieste, Italy 10:55-11:15 Model Reduction in Physics-Based Sound Synthesis Doug L. James and Timothy Langlois, Cornell University, USA 11:20-11:40 Reduced-order Models using Dynamic Mode Decomposition Clarence Rowley, Matthew O. Williams, Maziar S. Hemati, and Scott Dawson, Princeton University, USA 11:45-12:05 Iterative Solution Techniques in Reduced-order Modeling Virginia Forstall and Howard C. Elman, University of Maryland, College Park, USA 12:10-12:30 Online Adaptive Model Reduction Benjamin Peherstorfer and Karen E. Willcox, Massachusetts Institute of Technology, USA
162
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
MS261 Parallel Methods for Time Integration - Part I of II 10:55 AM-12:35 PM Room:250 E For Part 2 see MS286 Current trends in supercomputing are leading towards systems with more, but not faster, processors. Therefore, faster compute speeds must come from greater parallelism and this leads to a bottleneck for sequential time integration. Given this situation, interest in achieving parallelism in time has substantially increased in the past decade. A goal of this minisymposium is to survey the breadth of this emerging field and to detail recent advances. Eight talks are featured which bring together experts on four of the most popular current approaches: parareal, spectral deferred correction, multigrid reduction in time and the asymptotic parallel-in-time approach.
11:45-12:05 Parareal Library for Time-Dependent PDEs in Medical Applications Andreas Kreienbuehl, University of Lugano, Switzerland; Arne Naegel, Goethe University Frankfurt, Germany; Daniel Ruprecht, Universita’ della Svizzera Italiana, Italy; Robert Speck, Jülich Supercomputing Centre, Germany; Gabriel Wittum, Goethe University Frankfurt, Germany; Rolf Krause, University of Lugano, Switzerland 12:10-12:30 A Posteriori Analysis of the Parareal Algorithm: Efficient Resource Allocation and Convergence Criteria Jehanzeb H. Chaudhry, Florida State University, USA; Don Estep and Simon Tavener, Colorado State University, USA
Wednesday, March 18
MS262 Topology Mapping and Locality - Part I of II 10:55 AM-12:35 PM Room:250 F For Part 2 see MS287 This two part minisymposium focuses on topology mapping to improve locality and computational performance, an important issue that will become crucial as system size continues to increase. Specifically we look at the assignment of tasks to allocated processors for an application. In this first part, speakers will present results for algorithms to improve geometric task mapping, locality for sparse unstructured communication patterns, application performance using dynamic monitoring, and data management. In the second part, speakers will present results for graph models and dragonfly networks.
Organizer: Vitus Leung
Organizer: Jacob B. Schroder
Sandia National Laboratories, USA
Lawrence Livermore National Laboratory, USA
Organizer: Torsten Hoefler
Organizer: Ronald Haynes
10:55-11:15 Local Search to Improve Geometric Task Mapping Vitus Leung, Sandia National Laboratories, USA; David Bunde, Knox College, USA
Memorial University, Newfoundland, Canada
Organizer: Scott Maclachlan Tufts University, USA 10:55-11:15 An Overview of the Multigrid Reduction in Time (MGRIT) Method Stephanie Friedhoff, Katholieke Universiteit Leuven, Belgium 11:20-11:40 Multigrid Reduction in Time (MGRIT): A Flexible and NonIntrusive Method Jacob B. Schroder, Robert Falgout, Ulrike Meier Yang, Tzanio V. Kolev, and Veselin Dobrev, Lawrence Livermore National Laboratory, USA; Stephanie Friedhoff, Katholieke Universiteit Leuven, Belgium; Scott Maclachlan, Tufts University, USA
continued in next column
University of Illinois, USA
11:20-11:40 Demonstrating Improved Application Performance Using Dynamic Monitoring and Task Mapping Ann Gentile, James Brandt, Karen D. Devine, and Kevin Pedretti, Sandia National Laboratories, USA 11:45-12:05 Locality for Sparse Unstructured Communication Patterns Ozan Tuncer, Boston University, USA; Vitus Leung, Sandia National Laboratories, USA; Ayse Coskun, Boston University, USA 12:10-12:30 Topology Aware Process Placement and Data Management Emmanuel Jeannot, INRIA, France
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS263
MS264
Hydrodynamics at Small Scales - Part I of II 10:55 AM-12:35 PM
Numerical Methods for High-Dimensional Stochastic and Parametric Problems - Part IV of V 10:55 AM-12:35 PM
Room:251 A For Part 2 see MS288 With the increased interest in nano- and micro-fluidics, as well as biological systems, it has become necessary to develop tools for hydrodynamic calculations at microscopic and mesoscopic scales. This minisymposium will focus on advances in multiscale numerical methods for simulating flows at mesoscopic scales. Of particular interest will be fluctuating hydrodynamics of complex fluids such as reactive mixtures, colloidal passive and active suspensions, and multiphase fluids. Issues to be discussed will include the inclusion of thermal fluctuations in analytical and computational models, as well as applications in the physical sciences, biology, and engineering.
Organizer: Aleksandar Donev Courant Institute of Mathematical Sciences, New York University, USA 10:55-11:15 Fluctuating Hydrodynamic from the Theory of Coarse-Graining Pep Español, Universidad Nacional de Educación a Distancia, Spain 11:20-11:40 Fluctuating Hydrodynamics of Reactive Multispecies Mixtures John B. Bell, Lawrence Berkeley National Laboratory, USA 11:45-12:05 Modeling MultiPhase Flow Using Fluctuating Hydrodynamics Anuj Chaudhri and John B. Bell, Lawrence Berkeley National Laboratory, USA; Alejandro Garcia, San Jose State University, USA; Aleksandar Donev, Courant Institute of Mathematical Sciences, New York University, USA 12:10-12:30 Temporal Integrators for Fluctuating Hydrodynamics Steven D. Delong, Eric Vanden-Eijnden, and Aleksandar Donev, Courant Institute of Mathematical Sciences, New York University, USA
Room:251 B For Part 3 see MS240 For Part 5 see MS289 Development of scalable numerical methods for the solution of problems with highdimensional stochastic or parametric inputs has been a subject of active research in computational sciences and engineering. This is motivated by the need to reduce the issue of curse-of-dimensionality, i.e., exponential increase of computational complexity, in predictive simulation of physical systems where accurate specification of governing laws entails a large number of parameters or stochastic variables. To this end, several novel approaches based on multi-level, reduced order, sparse, and lowrank approximations have been recently developed. This minisymposium presents state-of- the-art in such developments for various aspects of high-dimensional computation, including analysis, algorithms, implementation, and applications.
Organizer: Dongbin Xiu University of Utah, USA
Organizer: Alireza Doostan University of Colorado Boulder, USA 10:55-11:15 A Low-Rank Approximation Method for High-Dimensional Uncertainty Quantification Alireza Doostan, University of Colorado Boulder, USA; Dongbin Xiu, University of Utah, USA; Akil Narayan, University of Massachusetts, Dartmouth, USA; Hillary Fairbanks, University of Colorado Boulder, USA 11:20-11:40 Bayesian Compressive Sensing Framework for HighDimensional Surrogate Construction Khachik Sargsyan, Cosmin Safta, Bert J. Debusschere, and Habib N. Najm, Sandia National Laboratories, USA
continued in next column
163
11:45-12:05 Sample-based Lowrank Methods for Tensor-structured Parametric Equations Loïc Giraldi, Anthony Nouy, and Olivier Zahm, Ecole Centrale de Nantes, France 12:10-12:30 Inverse Subspace Iteration for Spectral Stochastic Finite Element Methods Bedrich Sousedik, University of Maryland, Baltimore County, USA; Howard C. Elman, University of Maryland, College Park, USA
164
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS266
MS267
A New Class of Finite Element Methods: Weak Galerkin Methods Part I of II 10:55 AM-12:35 PM
Randomized Algorithms in Numerical Linear Algebra Part I of II 10:55 AM-12:35 PM
Inverse Problems and Data Assimilation - Part V of VI 10:55 AM-12:35 PM
Room:251 D
Room:251 C
For Part 2 see MS291 Recent work on developing randomized matrix algorithms has led to the highquality numerical implementations that can outperform deterministic methods while still providing very accurate results. This minisymposium will present randomized algorithms and software that enhance numerical linear algebra computations. The speakers will describe how randomized algorithms can accelerate general or structured matrix computations. The applications will include the computation of low-rank approximations, preconditioning for stochastic gradient algorithms, principal component analysis, random butterfly transformations in sparse linear systems, randomized HSS compressions and fast generation of random matrices.
For Part 4 see MS243 For Part 6 see MS292 The aim of this minisymposium is to document recent mathematical developments in the field of inverse problems and data assimilation based on theoretical and numerical grounds that are relevant for various scientific and real life applications. Topics for the minisymposium include but are not limited to: 1) deterministic and statistical approaches to inversion problems and data assimilation; 2) reduced order modeling inversion framework; 3) treatment of nonlinear and non-smooth processes; 4) sensitivity analysis and its application to adaptive (targeting) observation; 5) operational data assimilation systems; 6) uncertainties impact studies; 7) automatic tools to support inversion and data assimilation methodologies.
Organizer: Marc Baboulin
Organizer: Razvan Stefanescu
Wednesday, March 18
MS265
For Part 2 see MS290 A new class of finite element methods, called weak Galerkin (WG) methods, were recently introduced by Wang and Ye for some model PDEs with appropriately defined variational formulations. The WGFEMs, by design, make use of discontinuous piecewise polynomials on finite element partitions with arbitrary shape of polygons and polyhedrons. The WGFEMs have the flexibility in handling complex geometry and low regularity solutions, the simplicity in analyzing real-world physical problems and the symmetry in reformulating the original PDEs. The minisymposium aims to bring together researchers to exchange ideas on development of WGFEMs and its applications. The minisymposium pays a particular attention to attract female participants, graduate students, postdocs and junior faculty members.
Organizer: James Liu Colorado State University, USA 10:55-11:15 Weak Galerkin Methods for Partial Differential Equations Xiu Ye, University of Arkansas at Little Rock, USA 11:20-11:40 Computational Aspects of Weak Galerkin Methods Lin Mu, Michigan State University, USA 11:45-12:05 Weak Galerkin Methods for Linear Elasticity Problems Shangyou Zhang, University of Delaware, USA 12:10-12:30 Title Not Available at Time of Publication Mark Burg, Texas A&M University, USA
Room:251 E
INRIA/University of Paris-Sud, France
Virginia Tech, USA
Organizer: Jack J. Dongarra
Organizer: Adrian Sandu
University of Tennessee, Knoxville, USA
Virginia Tech, USA
Organizer: Xiaoye Sherry Li
Organizer: Ionel M. Navon
Lawrence Berkeley National Laboratory, USA
Florida State University, USA
Organizer: Dacian N. Daescu
10:55-11:15 Online Principal Component Analysis Christos Boutsidis, Yahoo! Labs, Santa Clara, USA
Portland State University, USA
11:20-11:40 Implementation of a Fast Multifrontal Solver Using Randomized HSS Compression Pieter Ghysels, Francois-Henry Rouet, and Xiaoye Sherry Li, Lawrence Berkeley National Laboratory, USA 11:45-12:05 Deterministic and Randomized CUR and Nystrom Approximations Ilse Ipsen, North Carolina State University, USA 12:10-12:30 Fast Generation of Random Orthogonal Matrices Amal Khabou, Francoise Tisseur, and Nicholas J. Higham, University of Manchester, United Kingdom
10:55-11:15 Model Reduction and Sensor Placement in a Feedback Flow Control Problem Jeff Borggaard, Serkan Gugercin, and Lizette Zietsman, Virginia Tech, USA 11:20-11:40 Goal-Based Rom Adjoint for Optimal Sensor Locations and Data Assimilation Fangxin Fang and Christopher Pain, Imperial College London, United Kingdom; Ionel M. Navon, Florida State University, USA; Zhizhao Che, Andrew G. Buchan, Pavlidis Dimitrios, and Dunhui Xiao, Imperial College London, United Kingdom
continued on next page
2015 SIAM Conference on Computational Science and Engineering
11:45-12:05 Reduced Order Modelling (rom) of the Navier-Stokes Equations for 3D Free Surface Flows Dunhui Xiao, Fangxin Fang, Christopher Pain, and Andrew G. Buchan, Imperial College London, United Kingdom; Ionel M. Navon, Florida State University, USA 12:10-12:30 Optimized Reduced Order Modeling and Data Assimilation for Hydrodynamics with Large Time Step Observations Diana Bistrian, Politechnic University of Timisoara, Romania; Ionel M. Navon, Florida State University, USA
Wednesday, March 18
MS268 Mathematical and Physical Properties of Numerical Schemes for Complex Dynamical Systems Part I of II 10:55 AM-12:35 PM Room:251 F For Part 2 see MS293 Stability and convergence are fundamental issues in the study of numerical algorithms. Another important aspect is their capability of mimicking certain physical properties of the target continuous system, such as energy conversion, or tracer transport. In this minisymposium, researchers from different areas in applied and computational mathematics are invited to present their latest results in studies of numerical schemes for complex dynamical system. Finite difference and finite volume schemes are of primary interests here, but finite element methods may also be included. Application areas include but are not limited to aeroelasticity, classical fluid dynamics, and geophysical fluid dynamics.
Organizer: Qingshan Chen Clemson University, USA
Organizer: Gung-Min Gie University of Louisville, USA 10:55-11:15 Numerical Weather Prediction in Two Dimensions with Topography, using a Finite Volume Method Roger M. Temam and Arthur Bousquet, Indiana University, USA; Mickael Chekroun, University of California, Los Angeles, USA; YoungJoon Hong, Indiana University, USA; Joseph J. Tribbia, National Center for Atmospheric Research, USA
continued in next column
165
11:20-11:40 A New Atmospheric Dynamic Core using 4th Order Flux Reconstruction Method with WENO Limiting Xingliang Li, China Meteorological Administration, China; Ziyao Sun and Feng Xiao, Tokyo Institute of Technology, Japan; Chungang Chen, Xi’an Jiaotong University, P.R. China; Xueshun Shen, China Meteorological Administration, China; Ming Xue, University of Oklahoma, USA 11:45-12:05 Higher Order Finite Volume Approximations of the Inviscid Primitive Equations in a Complex Domain Gung-Min Gie, University of Louisville, USA; Arthur Bousquet, YoungJoon Hong, and Roger M. Temam, Indiana University, USA 12:10-12:30 Title Not Available at Time of Publication Yau Shu Wong, University of Alberta, Canada
166
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS270
MS271
Flooding the Cores Computing Flooding Events with Many-Core and Accelerator Technologies Part I of II 10:55 AM-12:35 PM
Advances in Computational Methods for MHD and MultiFluid Models of Plasma Part I of II 10:55 AM-12:35 PM
Room:254 B
Organizer: Todd Munson
For Part 2 see MS294 Detailed simulation of flooding events, such as in storm surges, tsunamis or rain floods, requires substantial computing power and literally floods the computer with job tasks. Relevant supercomputing technology, on the other hand, is characterized by an increasing amount of parallelism on all scales. Hence, numerical methods, algorithms and software need to be tuned for data parallelism, manycore compute nodes (including accelerator technology) and large-scale parallelism. This minisymposium thus focuses on approaches to reduce time to solution for simulating complex flooding events on supercomputers and architectures characterized by many-core and accelerator technologies, in particular GPUs and Xeon Phis.
For Part 2 see MS295 Plasmas are ionized gases that appear in a wide range of applications including astrophysics and space physics, as well as in laboratory settings such as in magnetically confined fusion. Modeling and understanding the basic phenomenon in plasma have long been topics in applied mathematics, yet many problems remain far too numerically intensive for modern parallel computers. The main difficulty is that plasmas span a wide range of spatial and temporal scales, requiring a wide array of computational mathematics tools. This minisymposium aims to describe recent advances in the development of numerical methods and computational frameworks for the numerical solution of the magnetohydrodynamics (MHD) equations and other multi-fluid plasma physics models.
Argonne National Laboratory, USA
Organizer: Kyle T. Mandli
Organizer: James A. Rossmanith
10:55-11:15 Decomposition Algorithms for Transmission and Generation Investment Planning Francisco Munoz and Jean-Paul Watson, Sandia National Laboratories, USA
Columbia University, USA
Iowa State University, USA
Organizer: Michael Bader
Organizer: Lucian Ivan
Wednesday, March 18
MS269 Optimization Algorithms for Power Grid Expansion, Transmission, and Contingency Analysis 10:55 AM-12:35 PM Room:254 A Next generation power grid requires new mathematics and tailored optimization algorithms. This minisymposium covers four different aspects of decision-making process in power grid, namely, the expansion investment, economic dispatch, transmission scheduling, and contingency analysis. We bring together experts with the goal of presenting recent advances in these areas. Talks are organized with the emphasis on the role of computational tools and optimization algorithms in the development of next generation power grid.
Organizer: Fu Lin Argonne National Laboratory, USA
Organizer: Sven Leyffer Argonne National Laboratory, USA
11:20-11:40 Moment-Based Relaxations of the Optimal Power Flow Problem Daniel Molzahn and Ian Hiskens, University of Michigan, USA 11:45-12:05 Computational Study of Security-Constrained Economic Dispatch with Multi-Stage Rescheduling Yanchao Liu, University of Wisconsin, Madison, USA; Michael C. Ferris, University of Wisconsin, USA; Feng Zhao, ISO New England, USA 12:10-12:30 Towards Efficient N-x Contingency Selection Using Group Betweenness Centrality Mahantesh Halappanavar, Yousu Chen, and Zhenyu Huang, Pacific Northwest National Laboratory, USA
Room:254 C
TU München, Germany
University of Waterloo, Canada
Organizer: Tobias Weinzierl
Organizer: Qi Tang
Durham University, United Kingdom
Michigan State University, USA
10:55-11:15 Parallelization Techniques for Tsunami Simulation Using SpaceFillig-Curve Orders Michael Bader and Oliver Meister, TU München, Germany
10:55-11:15 Globally DivergenceFree Projection Methods for Ideal Magnetohydrodynamics James A. Rossmanith, Iowa State University, USA
11:20-11:40 QuickSched - Using Tasks for Massively Parallel Hybrid Shared/ distributed Memory Computing Pedro Gonnet, University of Oxford, United Kingdom
11:20-11:40 Block Adaptive MHD Simulations for Solar Coronal Dynamics Rony Keppens, Katholieke Universiteit Leuven, Belgium
11:45-12:05 Understanding Tsunami and Hurricane Deposits with a MessScale Model for Sediment Dynamics Robert Weiss and Wei Cheng, Virginia Tech, USA
11:45-12:05 Positivity-Preserving Weno Schemes with Constrained Transport for Ideal MHD Qi Tang, Michigan State University, USA
12:10-12:30 A Patchwork Family Task Distribution Patterns for Shallow Water Equations on Patch-structured AMR Grids Tobias Weinzierl, Durham University, United Kingdom
12:10-12:30 Multi-Fluid Plasma Modeling Through the Collisional Transition Regime Uri Shumlak, Andrew Ho, Robert Lilly, Sean Miller, Noah F. Reddell, and Eder Sousa, University of Washington, USA
2015 SIAM Conference on Computational Science and Engineering
167
Wednesday, March 18
Wednesday, March 18
Wednesday, March 18
MS272
MS273
MS274
Recent Advances in High Order Finite Element Methods for Atmospheric Sciences - Part I of II 10:55 AM-12:35 PM
Algorithms for the Eigenvalue Problem in Electronic Structure Computations - Part I of II 10:55 AM-12:35 PM
Models and Algorithms for Engineering Optimization Under Uncertainty Part I of II 10:55 AM-12:35 PM
Room:150 AB
Room:150 DE
Room:258
For Part 2 see MS296 Global and regional models are heavily used for weather forecasting, policy decisions and as compact laboratories for improving our scientific understanding of the Earth system. The next generation of numerical methods will soon be used for answering pressing questions on global/regional scale interactions, extreme events, and regional scale climate change. This minisymposium focuses on the latest advanced developments in high(er) order finite element methods including Discontinuous Galerkin,and high order finite volume methods. The speakers will address theoretical and computational issues such as stability, optimal order convergence, sparse discretization, parallel implementation, (hp)-adaptivity, large-scale problems and efficient implementations.
For Part 2 see MS297 The electronic structure of complex quantum mechanical systems is determined by the solution of a large number of one-dimensional non-linearly coupled Schrodinger equations. Upon discretization this set of equations translates into Hermitian symmetric eigenvalue problems. Due to the many existing discretization schemes eigenproblems come in many flavours and their solution requires a multidisciplinary approach. This minisymposium addresses the eigenproblems heterogeneity in connection with the rich variety of algorithms and implementations which are used to solve them. The ultimate goal is to make computational physicists and computer scientists aware of the current status of research and scientific advancement.
For Part 2 see MS298 Engineering optimization problems are often constrained by large-scale simulations of physical systems. In many applications, the parameters and inputs characterizing the physical system are unknown or estimated from data. When solving such optimization problems, it is not only important to accurately characterize the uncertainty and incorporate it in the optimization formulation, but to also determine optimal solutions that are robust or risk-averse to this uncertainty. This minisymposium presents novel risk models and efficient algorithms for the treatment of uncertainty in engineering optimization problems.
Organizer: Paul Ullrich
Organizer: Edoardo A. Di Napoli
Sandia National Laboratories, USA
Jülich Supercomputing Centre, Germany
University of California, Davis, USA
Organizer: Tan Bui-Thanh University of Texas at Austin, USA 10:55-11:15 Tempest: Efficient Computation of Atmospheric Flows Using High-Order Local Discretization Methods Paul Ullrich, University of California, Davis, USA 11:20-11:40 Vertical Discretization of Geophysical Flows with the Hybrid Finite Element Method Normal Mode and Wave Dispersion Properties Jorge E. Guerra and Paul Ullrich, University of California, Davis, USA 11:45-12:05 Toward Exa-Scale Computing in CAM-SE David M. Hall and Henry Tufo, University of Colorado Boulder, USA 12:10-12:30 A High-Order Global Discontinuous Galerkin NonHydrostatic Atmospheric Model Using Hevi Time Integration Scheme Ram Nair, National Center for Atmospheric Research, USA; Lei Bao, University of Colorado Boulder, USA
Organizer: Eric Polizzi University of Massachusetts, Amherst, USA
Organizer: Yousef Saad University of Minnesota, USA 10:55-11:15 Planning the Next Generation of Electronic Structure Codes James R. Chelikowsky, University of Texas at Austin, USA 11:20-11:40 Updating Strategies for Efficient Large-Scale Electronic Structure Calculations Roland Wittmann and Thomas K. Huckle, Technische Universität München, Germany 11:45-12:05 A Projected Preconditioned Conjugate Gradient Algorithm for Eigenvalue Calculation Chao Yang and Eugene Vecharynski, Lawrence Berkeley National Laboratory, USA; John Pask, Lawrence Livermore National Laboratory, USA 12:10-12:30 Ongoing Developments in BigDFT towards the ab-initio Computation of Resonant States Alessandro Cerioni, Université Grenoble, France; Luigi Genovese, Thierry Deutsch, Ivan Duchemin, and Maxime Moriniere, CEA, France
Organizer: Bart G. Van Bloemen Waanders Organizer: Drew P. Kouri Sandia National Laboratories, USA 10:55-11:15 Integration of Approximate Schur Preconditioners and SQP Algorithms for Nonlinear PDE Optimization under Uncertainty Denis Ridzal, Drew P. Kouri, and Bart G. Van Bloemen Waanders, Sandia National Laboratories, USA 11:20-11:40 Stochastic ReducedOrder Models in Optimization and Inverse Problems Wilkins Aquino, Duke University, USA; James Warner, NASA Langley Research Center, USA; Mircea Grigoriu, Cornell University, USA 11:45-12:05 Scalable Algorithms for Optimal Control of Systems Governed by PDEs under Uncertainty Alen Alexanderian, University of Texas at Austin, USA; Noemi Petra, University of California, Merced, USA; Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA; Omar Ghattas, University of Texas at Austin, USA
continued on next page
168
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS274
MS275
MS276
Models and Algorithms for Engineering Optimization Under Uncertainty Part I of II
Preconditioners for Sparse Least Squares 10:55 AM-12:35 PM Room:259
Advances in Compressed Sensing and Structured Sparse Representations Part I of II 10:55 AM-12:35 PM
Wednesday, March 18
continued
12:10-12:30 A Scalable Compositional Approach to Uncertainty Quantification for the Optimization under Uncertainty of Multi-physics Systems Doug Allaire, Texas A&M University, USA; Sergio Amaral, Massachusetts Institute of Technology, USA; Kaiyu Li, Texas A&M University, USA; Karen E. Willcox, Massachusetts Institute of Technology, USA
It’s natural to extend concepts and techniques used to solve sparse linear systems to solving linear least squares problems. This minisymposium brings together interested researchers with a variety of approaches. For the overdetermined case, we use iteration methods, e.g. conjugate gradient, improving conditioning of the iterative system by LU factorization and partial orthogonalization, tolerance for which depends on estimated conditioning of L. For the over- and underdetermined case, Krylov methods and inner iteration are successfully employed. Recursive multilevel techniques can improve efficiency. Finally, we explore some relevant theory of the conjugate gradient method.
Organizer: Gary W. Howell North Carolina State University, USA 10:55-11:15 A Recursive Multilevel Approximate Inverse-Based Preconditioner for Solving General Linear Systems Bruno Carpentieri, University of Groningen, Netherlands 11:20-11:40 Krylov Subspace Methods Preconditioned by Inner Iterations for Rank-Deficient Least Squares Problems Keiichi Morikuni and Miroslav Rozloznik, Czech Academy of Sciences, Czech Republic; Ken Hayami, National Institute of Informatics, Japan 11:45-12:05 Global Adaptive Dropping in Incomplete Factorizations Miroslav Tuma, Academy of Sciences of the Czech Republic, Prague, Czech Republic; Jiri Kopal, Technical University of Liberec, Czech Republic; Miro Rozloznik, Academy of Sciences of the Czech Republic, Prague, Czech Republic 12:10-12:30 Recursive Multilevel Approximate Inverse-Based Preconditioning Yiming Bu, University of Electronic Science and Technology of China, China, and University of Groningen, Netherlands; Bruno Carpentieri, University of Groningen, Netherlands
Room:260 A For Part 2 see MS300 The goal of this minisymposium is to present recent progress on models and algorithms for compressed sensing that go beyond the traditional sparsity setting. It has long been acknowledged that many practical applications of compressed sensing - ranging from medical sciences to optical imaging and wireless communications - possess substantially more structure than sparsity alone. Leveraging such structure presents significant challenges in both the construction of measurements (e.g. using tools from communications and optimization) and the design of new, often probabilistic, algorithms. This minisymposium will bring together leading experts in the field to latest research into these vital questions.
Organizer: Ben Adcock Simon Fraser University, Canada
Organizer: Anders Hansen University of Cambridge, United Kingdom 10:55-11:15 Statistical Methods in Compressive Sensing: Theory and Experiment Larry Carin, Duke University, USA 11:20-11:40 Model-Based Sketching and Recovery with Expanders Luca Baldassarre and Volkan Cevher, École Polytechnique Fédérale de Lausanne, Switzerland 11:45-12:05 Performance Limits of Ideal Decoders in Linear Inverse Problems Anthony Bourrier, Gipsa-Lab, France 12:10-12:30 Practical Compressed Sensing: On Asymptotic Structure Bogdan Roman, University of Cambridge, United Kingdom
2015 SIAM Conference on Computational Science and Engineering
169
Wednesday, March 18
Wednesday, March 18
Wednesday, March 18
MS277
MS278
MS279
Computational Techniques for Wave Equations in Second Order Form Part I of II 10:55 AM-12:35 PM
Advances in Time-Domain Boundary Integral Equations - Part I of II 10:55 AM-12:35 PM
Co-Design with Proxy Applications: Results and Experiences - Part I of II 10:55 AM-12:35 PM
Room:151 DE
Room:151 G
Room:260 B
For Part 2 see MS303 In recent years there has been an increasing interest in understanding and enhancing the simulation of linear evolution equations (transient waves, Stokes flow, and diffusion) using boundary integral equation (BIE) methods. This minisymposium brings together researchers in the field of timedomain BIE to enhance relationships and to discuss current progress and future trends in theory, computation and advanced applications. A very wide range of topics are covered such as fast methods, stable timestepping strategies (marching-on-in-time and convolution quadrature), well- posedness, and new applications in electromagnetism, acoustics, elastodynamics, and heat diffusion.
For Part 2 see MS304 Effective use of computing environments for scientific and engineering applications is determined by a combination issues throughout the codesign space: hardware, runtime environment, programming models, languages and compilers, algorithm choice and implementation, and more. Our focus is on applications that are large and complex, applying multi-physics at multi-scale, often with source code distribution constraints. Application proxies enable a language for codesign, providing a collaborative tool for exploring large-scale high performance scientific computation. Presentations in this minisymposium will describe experiences using proxies to explore key issues in computational science, providing examples across the codesign spectrum.
For Part 2 see MS301 Many wave systems are most naturally written in second order form. First order reformulations, though possible, typically require additional constraints and more variables. Despite this fact, the theory of efficient time-domain discretization schemes is better developed in the first order case. Speakers in this minisymposium will discuss the generalization of concepts such as upwinding which are familiar for first order systems, as well as more traditional methods which leverage special features of second order formulations. In addition, special techniques for treating high frequency waves and random media as well as applications to complex physical phenomena will be considered.
Organizer: Nicolas Salles
Organizer: Thomas M. Hagstrom
University College London, United Kingdom
Organizer: Richard Barrett
Southern Methodist University, USA
Organizer: Francisco J. J. Sayas
Sandia National Laboratories, USA
Organizer: Daniel Appelo
University of Delaware, USA
Organizer: Charles (Bert) H. Still
University of New Mexico, USA
10:55-11:15 An Exponentially Convergent Convolution Quadrature Method for Time-Domain Boundary Integral Equations Nicolas Salles and Timo Betcke, University College London, United Kingdom
Lawrence Livermore National Laboratory, USA
11:20-11:40 Adaptive Time Domain Boundary Element Methods (TD-BEM) for Scattering Problems Matthias Maischak and Matthias Gläfke, Brunel University, United Kingdom
11:20-11:40 Multi-Material ALE in the Blast Code Vladimir Tomov, Lawrence Livermore National Laboratory, USA
11:20-11:40 Performance Analysis of High-Order Discontinuous Galerkin Methods for First and Second Order Formulation of the Wave Equation Julien Diaz, INRIA Bordeaux Sud-Ouest, France 11:45-12:05 Second-Order Wave Equation with Uncertain Parameters: Analysis and Computation Mohammed Motamed, University of New Mexico, USA 12:10-12:30 Optimal Energy Conserving Local Discontinuous Galerkin Methods for Second-Order Wave Equation in Heterogeneous Media Ching-Shan Chou, The Ohio State University, USA
11:45-12:05 Fast Galerkin Method for Parabolic Space-Time Boundary Integral Equations Johannes Tausch, Southern Methodist University, USA 12:10-12:30 Convolution Quadrature Discretization of Volume Integral Equations Peter B. Monk, University of Delaware, USA
10:55-11:15 Co-Design Studies Using Mini-Multifluid-Ppm Paul R. Woodward, University of Minnesota, USA
Δ
10:55-11:15 The Double Absorbing Boundary Formulation of Complete Radiation Boundary Conditions Thomas M. Hagstrom, Southern Methodist University, USA
11:45-12:05 The -Nabla TimeComposite Approach for MultiPhysics Applications Productivity Jean-Sylvain Camier, CEA, France 12:10-12:30 Overview Co-Design at the DOE NNSA Trilabs Richard Barrett, Sandia National Laboratories, USA
Lunch Break 12:35 PM-2:00 PM Attendees on their own
170
Wednesday, March 18
MS280 Modeling Across the Curriculum 2:00 PM-3:40 PM Room:355 The second SIAM-NSF Workshop on Modeling across the curriculum (January, 2014) will be described, and an introduction to the report will be presented, together with some more recent progress. The meeting represented a more focused follow up to the 2012 workshop. The primary themes related to developing courses, programs, curricula, materials and training for a stronger emphasis on modeling and applied and computational mathematics in the early grades, middle and high school, and undergraduate programs. The intention is that presentations will be short to allow time for discussion.
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS281
MS282
Distributed Cyber-Physical Systems: Modelling and Controlling the Power Grid Part II of II 2:00 PM-3:40 PM
Recent Advances in Numerical Methods for Interface Problems Part IV of IV 2:00 PM-3:40 PM
Room:355 A
Room:250 A
For Part 1 see MS256 The Power Grid is a geographically distributed cyber-physical system, and its system complexity and the volume of data associated with it are growing. The speakers in the minisymposium will address recent advances in algorithms that accelerate the performance of software tools developed to model, optimize, and control the Grid.
For Part 3 see MS257 Many real world applications are characterized by multiple materials and complex fluids, and they often lead to interface problems. It is imperative to develop efficient and stable numerical methods for these problems. Great efforts have been made for solving interface problems and tracing the moving interfaces in the past decades. However, many challenges, such as the lack of regularity of physical solutions due to the discontinuity across interfaces, still hinder the development of efficient numerical methods. This minisymposium intends to create a forum for researchers from different fields to discuss recent advances on the interface methods and their applications.
Organizer: Alex Pothen Purdue University, USA
Organizer: Peter R. Turner
Organizer: Vaithianathan Venkatasubramanian
Clarkson University, USA
Washington State University, USA
2:00-2:20 Modeling Across the Curriculum: Introduction and Overview Peter R. Turner, Clarkson University, USA
Organizer: Mahantesh Halappanavar
2:25-2:45 Mathematical Modeling in the Early Grades Rachel Levy, Harvey Mudd College, USA 2:50-3:10 Applied and Computational Mathematics at the High School Level Katherine Socha, The Park School of Baltimore, USA; Kathleen Fowler, Clarkson University, USA 3:15-3:35 Modeling Across the Undergraduate Curriculum Jeffrey Humpherys, Brigham Young University, USA
Pacific Northwest National Laboratory, USA 2:00-2:20 Exploring State Estimation Techniques to Accommodate nonGaussian Noises Zhenyu Huang, Pacific Northwest National Laboratory, USA; Ning Zhou, Binghamton University, USA; Mihai Anitescu, Argonne National Laboratory, USA; Shaobu Wang, Pacific Northwest National Laboratory, USA 2:25-2:45 Efficient Algorithms for N-x Contingency Analysis for Power Grids Alex Pothen and Yu-Hong Yeung, Purdue University, USA; Mahantesh Halappanavar and Zhengyu Huang, Pacific Northwest National Laboratory, USA 2:50-3:10 Policy-switching Schemes for Power System Protection Rich Meier, Jesse Hostetler, Eduardo Cotilla-Sanchez, and Alan Fern, Oregon State University, USA 3:15-3:35 Singular Values and Convex Programming for Power System Synchrophasor Data Management Joe Chow and Meng Wang, Rensselaer Polytechnic Institute, USA
Organizer: Xiaoming He Missouri University of Science and Technology, USA
Organizer: Xiaolin Li State University of New York, Stony Brook, USA 2:00-2:20 Surfactant Driven Tipstreaming in a Flow Focusing Geometry Jacek Wrobel, Tulane University, USA; Michael Siegel and Michael R. Booty, New Jersey Institute of Technology, USA 2:25-2:45 Immersed Finite Element Methods with Enhanced Stability Xu Zhang, Purdue University, USA; Tao Lin, Virginia Tech, USA 2:50-3:10 A Finite Element Method for a Stokes Interface Problem Manuel A. Sanchez-Uribe, Brown University, USA 3:15-3:35 A Low-dimensional Approximation to the Stochastic Elliptic Interface Problem Ju Ming, Beijing Computational Science Research Center, China
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
MS283 Streamlining Application Performance Portability Part II of II 2:00 PM-3:40 PM Room:250 B For Part 1 see MS258 Achieving sustainable performance among different computer architectures often cannot be addressed with ingenuous combinations of compiler optimization options and manual rewriting. In contrast, auto-tuning (AT) technologies based, for example, on static modifications (typically pragmas), runtime optimizations, and source to source transformations have enabled sustained performance on a variety of computer architectures. As we move towards more complex computer architectures and higher levels of concurrency, AT is expected to play a major role in giving applications a gateway to performance portability. Of particular interest is the use of AT in the context of kernels that are the core of most large applications, including linear solvers, eigensolvers and stencil computations. In this minisymposium we will discuss AT frameworks and technologies, show examples of how they have been used in practice, and the corresponding performance gains.
Organizer: Toshiyuki Imamura RIKEN, Japan
Organizer: Takahiro Katagiri University of Tokyo, Japan
Organizer: Osni A. Marques Lawrence Berkeley National Laboratory, USA 2:00-2:20 Numerical Eigenvalue Engine towards Extreme-scale Computing Era Toshiyuki Imamura, Takeshi Fukaya, and Yusuke Hirota, RIKEN, Japan; Susumu Yamada and Masahiko Machida, Japan Atomic Energy Agency, Japan 2:25-2:45 Code Generation for Higher Level Spectral Methods with Spiral Franz Franchetti, Carnegie Mellon University, USA
continued in next column
2:50-3:10 Automatic Tuning for Parallel FFTs on GPU Clusters Daisuke Takahashi, University of Tsukuba, Japan 3:15-3:35 Statistical Performance Modeling and Autotuning for Dense QR Factorization in Hybrid CPU-GPU Systems Ray-Bing Chen, National Cheng Kung University, Taiwan; Yaohung Tsai and Weichung Wang, National Taiwan University, Taiwan
171
Wednesday, March 18
MS284 Parallel, Multiscale, Multiphysics Simulation Using MOOSE - Part II of II 2:00 PM-3:40 PM Room:250 C For Part 1 see MS259 The Idaho National Laboratory (INL)developed Multiphysics Object Oriented Simulation Environment (MOOSE; www. mooseframework.org), is an open-source, parallel computational platform that enables the solution of complex, fully-implicit multiphysics systems. MOOSE provides a set of computational tools that scientists and engineers can use to create sophisticated multiphysics simulations. Applications built using MOOSE have computed solutions for chemical reaction and transport equations, computational fluid dynamics, solid mechanics, heat conduction, mesoscale materials modeling, geomechanics, and others. This minisymposium highlights some of the recent results obtained using the MOOSE platform.
Organizer: Derek R. Gaston Idaho National Laboratory, USA
Organizer: Andrew Slaughter Idaho National Laboratory, USA 2:00-2:20 Modeling Nuclear Fuel Behavior with BISON Shane Stafford, Idaho National Laboratory, USA 2:25-2:45 Low Mach and TwoPhase Flow Modeling with Moose Applications Jean C. Ragusa and Marco Delchini, Texas A&M University, USA; Ray A. Berry, Idaho National Laboratory, USA 2:50-3:10 Multiphase Sub-Surface Flow Using Moose Jonathan Ennis-King, CSIRO, Australia 3:15-3:35 Stabilization Methods for High Peclet Number Flows in Heterogeneous Porous Media Yidong Xia, Hai Huang, and Robert Podgorney, Idaho National Laboratory, USA
172
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS286
MS287
Recent Advances in Model Reduction - Part V of V 2:00 PM-3:40 PM
Parallel Methods for Time Integration - Part II of II 2:00 PM-3:40 PM
Topology Mapping and Locality - Part II of II 2:00 PM-3:40 PM
Room:250 D
Room:250 E
Room:250 F
For Part 4 see MS260 Model reduction has become an increasingly important tool to mitigate the computational burden of modeling and simulation in timecritical (e.g., model-predictive control) and many query (e.g., Bayesian inversion) applications. This minisymposium presents recent advances that address the primary challenges facing such methods, such as preserving intrinsic problem structure, handling high- dimensional parameter spaces, integration with existing simulation codes, stability, optimization and uncertainty quantification and control.
For Part 1 see MS261 Current trends in supercomputing are leading towards systems with more, but not faster, processors. Therefore, faster compute speeds must come from greater parallelism and this leads to a bottleneck for sequential time integration. Given this situation, interest in achieving parallelism in time has substantially increased in the past decade. A goal of this minisymposium is to survey the breadth of this emerging field and to detail recent advances. Eight talks are featured which bring together experts on four of the most popular current approaches: parareal, spectral deferred correction, multigrid reduction in time and the asymptotic parallel-in-time approach.
For Part 1 see MS262 This two part minisymposium focuses on topology mapping to improve locality and computational performance, an important issue that will become crucial as system size continues to increase. Specifically we look at the assignment of tasks to allocated processors for an application. In this second part, speakers will present algorithms for graph models and dragonfly networks. In the first part, speakers will present results for algorithms to improve geometric task mapping, locality for sparse unstructured communication patterns, application performance using dynamic monitoring, and data management.
Organizer: Jacob B. Schroder
Sandia National Laboratories, USA
Wednesday, March 18
MS285
Organizer: Kevin T. Carlberg Sandia National Laboratories, USA
Organizer: Gianluigi Rozza SISSA, International School for Advanced Studies, Trieste, Italy 2:00-2:20 Reassessing the Missing Point Estimation Model Order Reduction Method Julien Cortial, Safran, Research & Technology Center, France; Kevin T. Carlberg, Sandia National Laboratories, USA 2:25-2:45 Parsimonious Data Acquisition for Data-driven Model Reduction Geoffrey M. Oxberry, Lawrence Livermore National Laboratory, USA 2:50-3:10 An Adaptive ParametrizedBackground Data-Weak Approach to State Estimation; Application to Heat Transfer Companion Experiments Tommaso Taddei, Masayuki Yano, James Penn, and Anthony T. Patera, Massachusetts Institute of Technology, USA 3:15-3:35 Reduced Basis Methods for Variational Inequalities Silke Glas, University of Ulm, Germany
Organizer: Vitus Leung
Lawrence Livermore National Laboratory, USA
Organizer: Torsten Hoefler
Organizer: Ronald Haynes
2:00-2:20 Process Mapping onto Complex Architectures and Partitions Thereof Francois Pellegrini, University of Bordeaux, France
Memorial University, Newfoundland, Canada
Organizer: Scott Maclachlan Tufts University, USA 2:00-2:20 Towards a Multigrid Perspective of MLSDC Dieter Moser and Robert Speck, Jülich Supercomputing Centre, Germany; Matthias Bolten, University of Wuppertal, Germany 2:25-2:45 An Adaptive Spectral Deferred Time Integrator for Vesicle Suspensions Bryan D. Quaife and George Biros, University of Texas at Austin, USA 2:50-3:10 Interweaving PFASST and Parallel Multigrid Matthias Bolten, University of Wuppertal, Germany; Michael Minion, Lawrence Berkeley National Laboratory, USA; Robert Speck, Jülich Supercomputing Centre, Germany; Matthew Emmett, Lawrence Berkeley National Laboratory, USA; Daniel Ruprecht, Universita’ della Svizzera Italiana, Italy 3:15-3:35 Parallel in Time Multigrid for Nonlinear Equations Ben O’Neill, University of Colorado Boulder, USA
University of Illinois, USA
2:25-2:45 Topology Aware Mapping using Graph Models for Exascale Systems Mehmet Deveci, The Ohio State University, USA; Kamer Kaya, CERFACS, France; Umit V. Catalyurek, The Ohio State University, USA 2:50-3:10 Maximizing Throughput on a Dragonfly Network Abhinav Bhatele, Lawrence Livermore National Laboratory, USA; Nikhil Jain, University of Illinois, USA; Xiang Ni and Laxmikant V Kale, University of Illinois at Urbana-Champaign, USA
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
MS288 Hydrodynamics at Small Scales - Part II of II 2:00 PM-3:40 PM Room:251 A For Part 1 see MS263 With the increased interest in nano- and micro-fluidics, as well as biological systems, it has become necessary to develop tools for hydrodynamic calculations at microscopic and mesoscopic scales. This minisymposium will focus on advances in multiscale numerical methods for simulating flows at mesoscopic scales. Of particular interest will be fluctuating hydrodynamics of complex fluids such as reactive mixtures, colloidal passive and active suspensions, and multiphase fluids. Issues to be discussed will include the inclusion of thermal fluctuations in analytical and computational models, as well as applications in the physical sciences, biology, and engineering.
Organizer: Aleksandar Donev Courant Institute of Mathematical Sciences, New York University, USA 2:00-2:20 Colloidal Dispersions Sheared at Constant Pressure John F. Brady, California Institute of Technology, USA 2:25-2:45 Fluctuating Hydrodynamics Methods for Soft Materials Paul J. Atzberger, University of California, Santa Barbara, USA 2:50-3:10 An Immersed Boundary Method for Rigid Bodies Bakytzhan Kallemov and Aleksandar Donev, Courant Institute of Mathematical Sciences, New York University, USA; Boyce Griffith, University of North Carolina at Chapel Hill, USA; Amneet Bhalla, Courant Institute of Mathematical Sciences, New York University, USA 3:15-3:35 A Fluctuating Immersed Boundary Method for Brownian Suspensions of Rigid Particles Aleksandar Donev and Bakytzhan Kallemov, Courant Institute of Mathematical Sciences, New York University, USA; Boyce Griffith, University of North Carolina at Chapel Hill, USA; Steven D. Delong, Courant Institute of Mathematical Sciences, New York University, USA
Wednesday, March 18
MS289 Numerical Methods for High-Dimensional Stochastic and Parametric Problems - Part V of V 2:00 PM-3:40 PM Room:251 B For Part 4 see MS264 Development of scalable numerical methods for the solution of problems with highdimensional stochastic or parametric inputs has been a subject of active research in computational sciences and engineering. This is motivated by the need to reduce the issue of curse-of-dimensionality, i.e., exponential increase of computational complexity, in predictive simulation of physical systems where accurate specification of governing laws entails a large number of parameters or stochastic variables. To this end, several novel approaches based on multi-level, reduced order, sparse, and lowrank approximations have been recently developed. This minisymposium presents state-of- the-art in such developments for various aspects of high-dimensional computation, including analysis, algorithms, implementation, and applications.
Organizer: Alireza Doostan University of Colorado Boulder, USA
Organizer: Dongbin Xiu University of Utah, USA 2:00-2:20 Gaussian Processes in HighDimensions Ilias Bilionis, Purdue University, USA; Nicholas Zabaras, Cornell University, USA 2:25-2:45 Adaptive Multivariate Interpolation Algorithm on Nested Grids and Its Application to Stochastic Collocation Xueyu Zhu, Yeonjong Shin, and Dongbin Xiu, University of Utah, USA 2:50-3:10 Numerical Solution for the High-Dimensional Joint ResponseExcitation Pdf Evolution Equations Heyrim Cho, Daniele Venturi, and George E. Karniadakis, Brown University, USA
continued in next column
173
3:15-3:35 High-Dimensional Hierarchical Uncertainty Quantification for Electronic Systems Zheng Zhang, Massachusetts Institute of Technology, USA; Xiu Yang, Pacific Northwest National Laboratory, USA; George E. Karniadakis, Brown University, USA; Ivan Oseledets, Skolkovo Institute of Science and Technology, Russia; Luca Daniel, Massachusetts Institute of Technology, USA
174
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS291
MS292
A New Class of Finite Element Methods: Weak Galerkin Methods Part II of II 2:00 PM-3:40 PM
Randomized Algorithms in Numerical Linear Algebra Part II of II 2:00 PM-3:40 PM
Inverse Problems and Data Assimilation - Part VI of VI 2:00 PM-3:40 PM
Room:251 D
Room:251 C
For Part 1 see MS266 Recent work on developing randomized matrix algorithms has led to the high-quality numerical implementations that can outperform deterministic methods while still providing very accurate results. This minisymposium will present randomized algorithms and software that enhance numerical linear algebra computations. The speakers will describe how randomized algorithms can accelerate general or structured matrix computations. The applications will include the computation of low-rank approximations, preconditioning for stochastic gradient algorithms, principal component analysis, random butterfly transformations in sparse linear systems, randomized HSS compressions and fast generation of random matrices.
For Part 5 see MS267 The aim of this minisymposium is to document recent mathematical developments in the field of inverse problems and data assimilation based on theoretical and numerical grounds that are relevant for various scientific and real life applications. Topics for the minisymposium include but are not limited to: 1) deterministic and statistical approaches to inversion problems and data assimilation;
Wednesday, March 18
MS290
For Part 1 see MS265 A new class of finite element methods, called weak Galerkin (WG) methods, were recently introduced by Wang and Ye for some model PDEs with appropriately defined variational formulations. The WGFEMs, by design, make use of discontinuous piecewise polynomials on finite element partitions with arbitrary shape of polygons and polyhedrons. The WGFEMs have the flexibility in handling complex geometry and low regularity solutions, the simplicity in analyzing real-world physical problems and the symmetry in reformulating the original PDEs. The minisymposium aims to bring together researchers to exchange ideas on development of WGFEMs and its applications. The minisymposium pays a particular attention to attract female participants, graduate students, postdocs and junior faculty members.
Organizer: James Liu Colorado State University, USA 2:00-2:20 Multiscale Weak Galerkin Methods Yalchin Efendiev, Texas A&M University, USA 2:25-2:45 Overview of Weak Galerkin Methods Junping Wang, National Science Foundation, USA 2:50-3:10 BDDC Domain Decomposition for Weak Galerkin Methods Xuemin Tu, University of Kansas, USA 3:15-3:35 Title Not Available at Time of Publication Yanqiu Wang, Oklahoma State University, USA
Organizer: Marc Baboulin
Room:251 E
2) reduced order modeling inversion framework; 3) treatment of nonlinear and non-smooth processes; 4) sensitivity analysis and its application to adaptive (targeting) observation; 5) operational data assimilation systems; 6) uncertainties impact studies;
INRIA/University of Paris-Sud, France
7) automatic tools to support inversion and data assimilation methodologies.
Organizer: Jack J. Dongarra
Organizer: Razvan Stefanescu
University of Tennessee, Knoxville, USA
Virginia Tech, USA
Organizer: Xiaoye Sherry Li
Organizer: Adrian Sandu
Lawrence Berkeley National Laboratory, USA
Virginia Tech, USA
2:00-2:20 Preconditioning Stochastic Gradient Algorithms with Randomized Linear Algebra Michael Mahoney, University of California, Berkeley, USA
Organizer: Ionel M. Navon
2:25-2:45 Randomized Methods for Accelerating Structured Matrix Computations Gunnar Martinsson, University of Colorado Boulder, USA
2:00-2:20 Challenges in Assimilation of PM2.5 Observations for Air Pollution Forecast Jiang JZhu, Chinese Academy of Sciences, China
2:50-3:10 Using Random Butterfly Transformations to Avoid Pivoting in Sparse Direct Methods Francois-Henry Rouet and Xiaoye Sherry Li, Lawrence Berkeley National Laboratory, USA; Marc Baboulin, INRIA/University of Paris-Sud, France
2:25-2:45 Inversion of Geothermal Heat Flux in a Thermomechanically Coupled Nonlinear Stokes Ice Sheet Model Hongyu Zhu, University of Texas at Austin, USA; Georg Stadler, Courant Institute of Mathematical Sciences, New York University, USA; Noemi Petra, University of California, Merced, USA; Toby Isaac, Omar Ghattas, and Thomas Hughes, University of Texas at Austin, USA
3:15-3:35 Performance of Computing Low-Rank Approximation on Hybrid CPU/GPU Architectures Ichitaro Yamazaki, University of Tennessee, Knoxville, USA; Theo Mary, Universite de Toulouse, France; Jakub Kurzak, Stanimire Tomov, and Jack J. Dongarra, University of Tennessee, Knoxville, USA
Florida State University, USA
Organizer: Dacian N. Daescu Portland State University, USA
continued on next page
2015 SIAM Conference on Computational Science and Engineering
2:50-3:10 Automated Adjoints for Mesh-Independent PDE-Constrained Optimisation Patrick E. Farrell, University of Oxford, United Kingdom 3:15-3:35 A Pod Model for Resolving the Angular Dimension of the Boltzmann Transport Equation Andrew G. Buchan, Atyab Calloo, Christopher Pain, Fangxin Fang, and Steven Dargaville, Imperial College London, United Kingdom; Ionel M. Navon, Florida State University, USA
175
Wednesday, March 18
Wednesday, March 18
MS293
MS294
Mathematical and Physical Properties of Numerical Schemes for Complex Dynamical Systems Part II of II 2:00 PM-3:40 PM
Flooding the Cores Computing Flooding Events with Many-Core and Accelerator Technologies Part II of II 2:00 PM-3:40 PM
Room:251 F
Room:254 B
For Part 1 see MS268 Stability and convergence are fundamental issues in the study of numerical algorithms. Another important aspect is their capability of mimicking certain physical properties of the target continuous system, such as energy conversion, or tracer transport. In this minisymposium, researchers from different areas in applied and computational mathematics are invited to present their latest results in studies of numerical schemes for complex dynamical system. Finite difference and finite volume schemes are of primary interests here, but finite element methods may also be included. Application areas include but are not limited to aeroelasticity, classical fluid dynamics, and geophysical fluid dynamics.
For Part 1 see MS270 Detailed simulation of flooding events, such as in storm surges, tsunamis or rain floods, requires substantial computing power and literally floods the computer with job tasks. Relevant supercomputing technology, on the other hand, is characterized by an increasing amount of parallelism on all scales. Hence, numerical methods, algorithms and software need to be tuned for data parallelism, manycore compute nodes (including accelerator technology) and large-scale parallelism. This minisymposium thus focuses on approaches to reduce time to solution for simulating complex flooding events on supercomputers and architectures characterized by many-core and accelerator technologies, in particular GPUs and Xeon Phis.
Organizer: Qingshan Chen
Organizer: Kyle T. Mandli
Clemson University, USA
Columbia University, USA
Organizer: Gung-Min Gie
Organizer: Michael Bader
University of Louisville, USA
TU München, Germany
2:00-2:20 The Effective Resolution of Advection Schemes James Kent, University of Michigan, USA; Jared P. Whitehead, Brigham Young University, USA; Christiane Jablonowski and Richard Rood, University of Michigan, USA
Organizer: Tobias Weinzierl
2:25-2:45 A New Adaptive Weighted Essentially Non-oscillatory WENO-Θ Scheme for Hypberbolic Conservation Laws Chang-Yeol Jung and Thien Binh Nguyen, Ulsan National Institute of Science and Technology, South Korea 2:50-3:10 Stability and Convergence of the Co-volume Scheme for the Stokes Problem Qingshan Chen, Clemson University, USA 3:15-3:35 Semi-Analytical Time Differencing Methods for Stiff Problems Chang-Yeol Jung and Thien Binh Nguyen, Ulsan National Institute of Science and Technology, South Korea
Durham University, United Kingdom 2:00-2:20 FEM Integration with Quadrature on the GPU Matthew Knepley, University of Chicago, USA; Karl Rupp and Karl Rupp, Vienna University of Technology, Austria; Andy R. Terrel, Continuum Analytics, USA 2:25-2:45 Flooding with Equelle: A Domain Specific Language for Finite Volume Methods Andre R. Brodtkorb, SINTEF, Norway 2:50-3:10 High Performance High Order Numerical Method for Tsunami Wave Propagation Rajesh Gandham, Timothy Warburton, and David Medina, Rice University, USA 3:15-3:35 Thermal Comfort Simulations on Massive Parallel Systems Ralf-Peter Mundani and Jérôme Frisch, Technische Universität München, Germany
176
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS296
MS297
Advances in Computational Methods for MHD and MultiFluid Models of Plasma Part II of II 2:00 PM-3:40 PM
Recent Advances in High Order Finite Element Methods for Atmospheric Sciences - Part II of II 2:00 PM-3:40 PM
Algorithms for the Eigenvalue Problem in Electronic Structure Computations - Part II of II 2:00 PM-3:40 PM
Room:254 C
Room:150 AB
Room:150 DE
For Part 1 see MS271 Plasmas are ionized gases that appear in a wide range of applications including astrophysics and space physics, as well as in laboratory settings such as in magnetically confined fusion. Modeling and understanding the basic phenomenon in plasma have long been topics in applied mathematics, yet many problems remain far too numerically intensive for modern parallel computers. The main difficulty is that plasmas span a wide range of spatial and temporal scales, requiring a wide array of computational mathematics tools. This minisymposium aims to describe recent advances in the development of numerical methods and computational frameworks for the numerical solution of the magnetohydrodynamics (MHD) equations and other multi-fluid plasma physics models.
For Part 1 see MS272 Global and regional models are heavily used for weather forecasting, policy decisions and as compact laboratories for improving our scientific understanding of the Earth system. The next generation of numerical methods will soon be used for answering pressing questions on global/regional scale interactions, extreme events, and regional scale climate change. This minisymposium focuses on the latest advanced developments in high(er) order finite element methods including Discontinuous Galerkin,and high order finite volume methods. The speakers will address theoretical and computational issues such as stability, optimal order convergence, sparse discretization, parallel implementation, (hp)-adaptivity, large-scale problems and efficient implementations.
For Part 1 see MS273 The electronic structure of complex quantum mechanical systems is determined by the solution of a large number of one-dimensional non-linearly coupled Schrodinger equations. Upon discretization this set of equations translates into Hermitian symmetric eigenvalue problems. Due to the many existing discretization schemes eigenproblems come in many flavours and their solution requires a multidisciplinary approach. This minisymposium addresses the eigenproblems heterogeneity in connection with the rich variety of algorithms and implementations which are used to solve them. The ultimate goal is to make computational physicists and computer scientists aware of the current status of research and scientific advancement.
Organizer: James A. Rossmanith
Organizer: Paul Ullrich
Jülich Supercomputing Centre, Germany
Iowa State University, USA
University of California, Davis, USA
Organizer: Lucian Ivan
Organizer: Tan Bui-Thanh
University of Waterloo, Canada
University of Texas at Austin, USA
Organizer: Qi Tang
2:00-2:20 An HDG Method for NonHydrostatic Atmosphere Tan Bui-Thanh, University of Texas at Austin, USA
Wednesday, March 18
MS295
Michigan State University, USA 2:00-2:20 A High-order Block-adaptive Simulation Framework for Ideal and Resistive MHD Equations on Cubedsphere Grids Lucian Ivan, Hans De Sterck, and Andree Susanto, University of Waterloo, Canada; Clinton P. Groth, University of Toronto, Canada 2:25-2:45 Scalable Solvers for Extended MHD in the Low-β Regime Luis Chacon, Los Alamos National Laboratory, USA 2:50-3:10 Block Preconditioners for 3D Incompressible MHD Eric C. Cyr, Edward Phillips, and John Shadid, Sandia National Laboratories, USA 3:15-3:35 Multi-Fluid Magnetohydrodynamic Models for Partially-Ionized Non-Equilibrium Anisotropic Plasmas Clinton P. Groth and Ken Miura, University of Toronto, Canada
2:25-2:45 A Higher-Order Finite Volume Nonhydrostatic Dynamical Core with Space-Time Refinement Hans Johansen, Lawrence Berkeley National Laboratory, USA 2:50-3:10 Towards a Fully 3D Compressible Atmosphere Dynamical Core with Compatible Finite Elements Colin J. Cotter, Imperial College London, United Kingdom 3:15-3:35 Optimization-based Spectral Element Semi-Lagrangian Tracer Transport Kara Peterson and Mark A. Taylor, Sandia National Laboratories, USA
Organizer: Edoardo A. Di Napoli Organizer: Eric Polizzi University of Massachusetts, Amherst, USA
Organizer: Yousef Saad University of Minnesota, USA 2:00-2:20 Accelerating Quantum Transport Calculations Through the Feast Algorithm Sascha Brueck, Mauro Calderara, Hossein Bani-Hashemian, Joost VandeVondele, and Mathieu Luisier, ETH Zürich, Switzerland 2:25-2:45 Parallel Solution of Eigenvalue Problems from Graphene Modeling with Solvers Based on Integration and Approximation Bruno Lang, Lukas Krämer, and Martin Galgon, University of Wuppertal, Germany 2:50-3:10 Polynomial Techniques and Primme for the Computation of Large Number of Eigenvalues Andreas Stathopoulos, College of William & Mary, USA 3:15-3:35 Linear Response Eigenvalue Problem and Excited State Calculations Zhaojun Bai, University of California, Davis, USA; Ren-Cang Li, University of Texas at Arlington, USA
2015 SIAM Conference on Computational Science and Engineering
177
Wednesday, March 18
Wednesday, March 18
Wednesday, March 18
MS298
MS299
MS300
Models and Algorithms for Engineering Optimization Under Uncertainty Part II of II 2:00 PM-3:40 PM
Cyberlearning Technology and Deep Learning Assessment in CSE Education 2:00 PM-3:40 PM
Advances in Compressed Sensing and Structured Sparse Representations Part II of II 2:00 PM-3:40 PM
Room:258
Room:259
Room:260 A
For Part 1 see MS274 Engineering optimization problems are often constrained by large-scale simulations of physical systems. In many applications, the parameters and inputs characterizing the physical system are unknown or estimated from data. When solving such optimization problems, it is not only important to accurately characterize the uncertainty and incorporate it in the optimization formulation, but to also determine optimal solutions that are robust or risk-averse to this uncertainty. This minisymposium presents novel risk models and efficient algorithms for the treatment of uncertainty in engineering optimization problems.
Computational science and engineering education is being transformed through a range of educational innovations. This minisymposium will comprise reports on some novel ideas to advance undergraduate CSE education through cyberlearning technology, deep learning assessment and innovative undergraduate research summer workshops.
For Part 1 see MS276 The goal of this minisymposium is to present recent progress on models and algorithms for compressed sensing that go beyond the traditional sparsity setting. It has long been acknowledged that many practical applications of compressed sensing - ranging from medical sciences to optical imaging and wireless communications - possess substantially more structure than sparsity alone. Leveraging such structure presents significant challenges in both the construction of measurements (e.g. using tools from communications and optimization) and the design of new, often probabilistic, algorithms. This minisymposium will bring together leading experts in the field to latest research into these vital questions.
Organizer: Bart G. Van Bloemen Waanders Sandia National Laboratories, USA
Organizer: Drew P. Kouri Sandia National Laboratories, USA 2:00-2:20 On Risk-averse PDEconstrained Optimization using Convex Risk Measures Inspired by Conditional Value-at-risk Thomas M. Surowiec, Humboldt University at Berlin, Germany 2:25-2:45 Maximizing AUC and Buffered AUC in Classification Stan Uryasev, University of Florida, USA 2:50-3:10 Optimization under Uncertainty: Application to Electrical Circuits Timur Takhtaganov, Rice University, USA 3:15-3:35 Optimal Control Problems With Uncertain Model Parameters Xiaodi Deng, Rice University, USA
Organizer: Hong Liu Embry-Riddle Aeronautical University, USA
Organizer: Michael Spector University of North Texas, USA 2:00-2:20 Tri-Located Course in Mathematical Modeling and Complementary Reu Summer Workshop Hong Liu and Andrei Ludu, Embry-Riddle Aeronautical University, USA 2:25-2:45 Deep Learning Assessment for Near Real-time, Formative Feedback during Complex Problemsolving Activities Michael Spector, University of North Texas, USA 2:50-3:10 Towards Automating Analysis of Midterm Semester Feedback Surveys for Improving Course Effectiveness Douglas Holton, Embry-Riddle Aeronautical University, USA 3:15-3:35 Facilitating Learners’ Cognitive Presence In A Self-Directed Online Course Ye Chen and Jing Lei, Syracuse University, USA
Organizer: Ben Adcock Simon Fraser University, Canada
Organizer: Anders Hansen University of Cambridge, United Kingdom 2:00-2:20 Title Not Available at Time of Publication Richard G. Baraniuk, Rice University, USA 2:25-2:45 Representation Using the Weyl Transform Robert Calderbank, Duke University, USA 2:50-3:10 Fast and Robust Dictionary Learning, with Invariances and Multiresolution Mauro Maggioni and Samuel Gerber, Duke University, USA 3:15-3:35 Compressive Parameter Estimation via Approximate Message Passing Shermin Hamzehei and Marco F. Duarte, University of Massachusetts, USA
178
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
Wednesday, March 18
MS301
MS302
MS303
Computational Techniques for Wave Equations in Second Order Form Part II of II 2:00 PM-3:40 PM
PDE-constrained Optimization using the Open-source Code SU2 2:00 PM-3:40 PM
Advances in Time-Domain Boundary Integral Equations - Part II of II 2:00 PM-3:40 PM
Room:151 AB
Room:151 DE
Room:260 B
Most established codes for PDE-constrained optimization are proprietary, unavailable, or prohibitively expensive for many users. The SU2 code is freely available as open-source and features a complete computational analysis framework for multidisciplinary design in applications such as, but not limited to, aerospace technology. This minisymposium will cover up-to-date topics within the SU2 framework related to its continuous and discrete adjoint capabilities, the application to large-scale aerodynamic design, and the utilization of many-core architectures. Each of the topics covered involve the combination of multiple research areas of interest to the CS&E community.
For Part 1 see MS278 In recent years there has been an increasing interest in understanding and enhancing the simulation of linear evolution equations (transient waves, Stokes flow, and diffusion) using boundary integral equation (BIE) methods. This minisymposium brings together researchers in the field of timedomain BIE to enhance relationships and to discuss current progress and future trends in theory, computation and advanced applications. A very wide range of topics are covered such as fast methods, stable timestepping strategies (marching-on-in-time and convolution quadrature), well- posedness, and new applications in electromagnetism, acoustics, elastodynamics, and heat diffusion.
Wednesday, March 18
For Part 1 see MS277 Many wave systems are most naturally written in second order form. First order reformulations, though possible, typically require additional constraints and more variables. Despite this fact, the theory of efficient time-domain discretization schemes is better developed in the first order case. Speakers in this minisymposium will discuss the generalization of concepts such as upwinding which are familiar for first order systems, as well as more traditional methods which leverage special features of second order formulations. In addition, special techniques for treating high frequency waves and random media as well as applications to complex physical phenomena will be considered.
Organizer: Thomas M. Hagstrom Southern Methodist University, USA
Organizer: Daniel Appelo University of New Mexico, USA 2:00-2:20 Upwind DG for Acoustic and Elastic Wave Equations Daniel Appelo, University of New Mexico, USA; Thomas M. Hagstrom, Southern Methodist University, USA 2:25-2:45 A Discontinuous Galerkin Method for the Spherically Reduced Einstein Field Equations with SecondOrder Operators Scott Field, Cornell University, USA; Jan Hesthaven, EPFL, France; Stephen Lau, University of New Mexico, USA; Abdul Mroue, Canadian Institute for Theoretical Astrophysics, Canada 2:50-3:10 High-Order Upwind Methods for Second-Order Wave Equations on Curvilinear and Overlapping Grids Jeffrey W. Banks, Rensselaer Polytechnic Institute, USA 3:15-3:35 Uncertainty Quantification for High Frequency Waves Olof Runborg, KTH Stockholm, Sweden
Organizer: Nicolas R. Gauger Technische Universität Kaiserslautern, Germany
Organizer: Juan J. Alonso Stanford University, USA 2:00-2:20 Large Scale Design Using Su2 and a Continuous Adjoint Rans Approach Francisco Palacios, Stanford University, USA 2:25-2:45 A Consistent and Robust Discrete Adjoint Solver for the SU2 Framework Tim Albring and Nicolas R. Gauger, Technische Universität Kaiserslautern, Germany 2:50-3:10 A Discrete Adjoint Framework for Lift-Constrained Noise Minimization Using SU2 Beckett Zhou and Nicolas R. Gauger, Technische Universität Kaiserslautern, Germany; Thomas Economon, Francisco Palacios, and Juan J. Alonso, Stanford University, USA 3:15-3:35 High-Performance Optimizations of the Unstructured Open-Source SU2 Suite Thomas Economon, Francisco Palacios, and Juan J. Alonso, Stanford University, USA; Gaurav Bansal, Anand Deshpande, Alexander Heinecke, Dheevatsa Mudigere, and Mikhail Smelyanskiy, Intel Corporation, USA
Organizer: Nicolas Salles University College London, United Kingdom
Organizer: Francisco J. J. Sayas University of Delaware, USA 2:00-2:20 Time-Domain Simulation of Two Dimensional Elastic Scattering Victor Dominguez, Universidad Pública de Navarra, Spain; Tonatiuh Sanchez-Vizuet and Francisco J. J. Sayas, University of Delaware, USA 2:25-2:45 Variable Order Fast Multipole Method for an Elastodynamic BEM Thomas Traub and Martin Schanz, Graz University of Technology, Austria 2:50-3:10 Recent Advances in the Convolution Quadrature and Temporal Galerkin Approaches to Transient Electromagnetics Daniel Weile, University of Delaware, USA; Balasubramaniam Shanker, Michigan State University, USA 3:15-3:35 Accuracy of the Marchingon-in-Time Scheme for Td-Bie Methods Elwin Van ‘t Wout, University College London, United Kingdom
2015 SIAM Conference on Computational Science and Engineering
Wednesday, March 18
MS304 Co-Design with Proxy Applications: Results and Experiences - Part II of II 2:00 PM-3:40 PM Room:151 G For Part 1 see MS279 Effective use of computing environments for scientific and engineering applications is determined by a combination issues throughout the codesign space: hardware, runtime environment, programming models, languages and compilers, algorithm choice and implementation, and more. Our focus is on applications that are large and complex, applying multi-physics at multi-scale, often with source code distribution constraints. Application proxies enable a language for codesign, providing a collaborative tool for exploring large-scale high performance scientific computation. Presentations in this minisymposium will describe experiences using proxies to explore key issues in computational science, providing examples across the codesign spectrum.
Organizer: Richard Barrett Sandia National Laboratories, USA
Organizer: Charles (Bert) H. Still Lawrence Livermore National Laboratory, USA 2:00-2:20 Experiences in Development of a Contact MiniApplication Using Kokkos Glen Hansen and Patrick Xavier, Sandia National Laboratories, USA 2:25-2:45 Co-Designing Hierarchical Algorithms: Application to VlasovMaxwell Particle-in-Cell Methods Joshua Payne, Luis Chacon, Guangye Chen, Chris Newman, Dana Knoll, and Allen McPherson, Los Alamos National Laboratory, USA 2:50-3:10 Uintah/Wasatch: Addressing Multiphsyics Complexity in a HighPerformance Computing Environment Tony Saad, Christopher Earl, Abhishek Bagusetty, Matthew Might, and James C. Sutherland, University of Utah, USA
Conference Adjourns 3:40 PM
179
180
2015 SIAM Conference on Computational Science and Engineering
Notes
2015 SIAM Conference on Computational Science and Engineering
Speaker Index
Italicized names are minisymposium organizers.
181
182
2015 SIAM Conference on Computational Science and Engineering
A
Amaya, Joseph, MS216, 2:15 Tue
Acar, Evrim, PD1, 12:15 Sat
Amir, Lubna, PP102, 4:30 Sun
Baboulin, Marc, MS266, 10:55 Wed
Amsallem, David, MS4, 10:15 Sat
Baboulin, Marc, MS291, 2:00 Wed
Amsallem, David, MS30, 2:25 Sat
Babtie, Ann C., MS132, 10:25 Mon
Amsallem, David, MS55, 4:35 Sat
Bacuta, Constantin, MS37, 3:15 Sat
Amsallem, David, MS187, 11:15 Tue
Bader, Michael, MS270, 10:55 Wed
Anderson, Jeffrey, MS243, 5:15 Tue
Bader, Michael, MS294, 2:00 Wed
Andini Putri, Fauziah, MS146, 10:14 Mon
Badia, Rosa M., MS171, 1:30 Mon
Andrews, Steven, MS90, 9:10 Sun
Adler, James H., MS85, 9:10 Sun
Badia, Santiago, MS32, 2:25 Sat
Anistratov, Dmitriy Y., MS223, 2:40 Tue
Adler, James H., MS85, 10:00 Sun
Badr, Abdallah A., PP13, 4:30 Mon
Anitescu, Mihai, MS195, 10:50 Tue
Adler, James H., MS110, 1:30 Sun
Baer, Steven M., MS153, 9:10 Mon
announced, To be, MS105, 10:50 Tue
Adler, James H., MS137, 9:10 Mon
Baer, Steven M., MS153, 9:10 Mon
Antil, Harbir, MS174, 1:55 Mon
Afkhami, Shahriar, MS245, 5:15 Tue
Bahlla, Upinder, MS115, 2:20 Sun
Anzt, Hartwig, PP205, 4:30 Mon
Aghakhani, Hossein, MS165, 2:20 Mon
Bai, Zhaojun, MS297, 3:15 Wed
Anzt, Hartwig, PP205, 4:30 Mon
Agullo, Emmanuel, MS82, 9:10 Sun
Bakhos, Tania, MS31, 3:15 Sat
Appelhans, David A., CP8, 10:10 Wed
Balay, Satish, PP103, 4:30 Sun
Agullo, Emmanuel, MS107, 1:30 Sun
Appelo, Daniel, MS36, 2:25 Sat
Baldassarre, Luca, MS276, 11:20 Wed
Agullo, Emmanuel, MS134, 9:10 Mon
Appelo, Daniel, MS277, 10:55 Wed
Ballani, Jonas, MS56, 5:50 Sat
Agullo, Emmanuel, MS145, 9:10 Mon
Appelo, Daniel, MS301, 2:00 Wed
Ballard, Grey, CP8, 9:25 Wed
Agullo, Emmanuel, MS160, 1:30 Mon
Appelo, Daniel, MS301, 2:00 Wed
Ballarin, Francesco, MS236, 5:40 Tue
Agullo, Emmanuel, MS171, 1:30 Mon
Aquino, Wilkins, MS274, 11:20 Wed
Banks, Jeffrey W., MS6, 10:15 Sat
Ahmadia, Aron, PP6, 4:30 Sun
Aravkin, Aleksandr, MS9, 10:15 Sat
Banks, Jeffrey W., MS32, 2:25 Sat
Ahmedov, Bahodir, PP4, 4:30 Sun
Aravkin, Aleksandr, MS9, 11:30 Sat
Banks, Jeffrey W., MS57, 4:35 Sat
Ahrens, James, IP9, 8:15 Wed
Aravkin, Aleksandr, MS35, 2:25 Sat
Banks, Jeffrey W., MS301, 2:50 Wed
Ajay, Aprant, MS94, 9:26 Sun
Arbenz, Peter, MS160, 1:30 Mon
Bao, Feng, PP201, 4:30 Mon
Albin, Nathan, MS3, 10:15 Sat
Archer, Cristina L., MS253, 4:50 Tue
Bao, Lei, MS272, 12:10 Wed
Albright, Jason, MS10, 10:15 Sat
Archibald, Richard, PP3, 4:30 Sun
Albring, Tim, MS302, 2:25 Wed
Arnold, Douglas N., PP207, 4:30 Mon
Barajas-Solano, David A., MS256, 12:10 Wed
Alexanderian, Alen, MS274, 11:45 Wed
Arrigo, Francesca, CP26, 9:10 Wed
Baraniuk, Richard G., MS2, 11:30 Sat
Alexeev, Alexander, MS42, 3:40 Sat
Baraniuk, Richard G., MS300, 2:00 Wed
Aljuhani, Shaimaa M., PP14, 4:30 Mon
Asante-Asamani, Emmanuel O., PP4, 4:30 Sun
Alkahtani, Badr, PP1, 4:30 Sun
Attia, Ahmed, MS218, 3:30 Tue
Barba, Lorena A., MS27, 2:25 Sat
Alla, Alessandro, MS174, 2:45 Mon
Atzberger, Paul J., PP104, 4:30 Sun
Barba, Lorena A., MS78, 10:25 Sun
Allaire, Doug, MS274, 12:10 Wed
Atzberger, Paul J., MS242, 4:25 Tue
Barba, Lorena A., PD3, 12:15 Mon
Alldredge, Graham, PP105, 4:30 Sun
Atzberger, Paul J., MS288, 2:25 Wed
Bardhan, Jaydeep, MS151, 9:35 Mon
Alldredge, Graham, MS147, 9:35 Mon
Audouze, Christophe, CP22, 9:40 Wed
Bardhan, Jaydeep, MS242, 5:15 Tue
Allen, Jeffery M., MS110, 1:55 Sun
Augustin, Florian, MS139, 10:25 Mon
Bardsley, Johnathan M., MS5, 10:15 Sat
Allmaras, Steven R., MS98, 9:35 Sun
Augustin, Florian, MS183, 10:00 Tue
Bardsley, Johnathan M., MS144, 9:10 Mon
Almgren, Ann S., MS201, 11:15 Tue
Auphan, Thomas, MS66, 5:00 Sat
Alolyan, Ibraheem, PP3, 4:30 Sun
Austin, Anthony, PP204, 4:30 Mon
Alonso, Juan J., MS302, 2:00 Wed
Austin, Brian, MS186, 10:50 Tue
Aluie, Hussein, CP24, 10:10 Wed
Austin, Woody N., CP25, 10:10 Wed
Amaral, Sergio, PP13, 4:30 Mon
Awanou, Gerard, PP207, 4:30 Mon
Adams, Mark, MS175, 1:55 Mon Adams, Mark, MS210, 3:30 Tue Adams, Marvin L., MS45, 2:25 Sat Adcock, Ben, MS276, 10:55 Wed Adcock, Ben, MS300, 2:00 Wed Adcroft, Alistair, MS157, 2:45 Mon Aderogba, Adebayo A., PP2, 4:30 Sun
Aminfar, Amirhossein, MS203, 10:50 Tue
Azijli, Iliass, PP14, 4:30 Mon
B
Babaee, Hessameddin, CP22, 9:25 Wed
Barba, Lorena A., MS27, 2:25 Sat
Bardsley, Johnathan M., MS170, 1:30 Mon Bardsley, Johnathan M., MS170, 1:30 Mon Barker, Andrew T., MS186, 10:25 Tue Barlow, Andrew J., MS156, 2:45 Mon Barlow, Jesse L., MS63, 5:00 Sat Barnett, Alex H., MS227, 2:15 Tue
2015 SIAM Conference on Computational Science and Engineering
183
Barra, Valeria, PP1, 4:30 Sun
Bertrand, Fleurianne, MS110, 2:20 Sun
Bochev, Pavel, PP203, 4:30 Mon
Barrett, Richard, MS279, 10:55 Wed
Berzins, Martin, MS93, 10:25 Sun
Bock, Nicolas, MS241, 5:40 Tue
Barrett, Richard, MS279, 12:10 Wed
Berzins, Martin, MS129, 9:10 Mon
Bodony, Daniel J., MS86, 9:10 Sun
Barrett, Richard, MS304, 2:00 Wed
Betcke, Timo, MS151, 9:10 Mon
Bodony, Daniel J., MS111, 1:30 Sun
Bartel, Andreas, MS162, 1:30 Mon
Betcke, Timo, MS151, 9:10 Mon
Bodony, Daniel J., MS111, 1:30 Sun
Bartlett, Roscoe, MS77, 5:00 Sat
Betcke, Timo, MS177, 1:30 Mon
Bokanowski, Olivier, MS15, 10:40 Sat
Bartol, Thomas M., MS90, 10:00 Sun
Betcke, Timo, MS203, 10:00 Tue
Bolten, Matthias, CP2, 9:10 Wed
Basting, Christopher, PP203, 4:30 Mon
Betcke, Timo, MS227, 2:15 Tue
Bolten, Matthias, MS286, 2:50 Wed
Basting, Christopher, CP14, 9:40 Wed
Bhaganagar, Kiran, MS253, 4:25 Tue
Boman, Erik G., MS141, 9:10 Mon
Bauer, Andrew, MS77, 5:50 Sat
Bhaganagar, Kiran, MS253, 4:25 Tue
Boman, Erik G., CP7, 10:10 Wed
Bauer, Martin, MS124, 1:30 Sun
Bhatele, Abhinav, MS287, 2:50 Wed
Borges, Carlos C., MS48, 3:40 Sat
Bauer, Martin, MS124, 1:30 Sun
Bhattacharya, Kaushik, MS14, 10:15 Sat
Borggaard, Jeff, MS148, 10:25 Mon
Bauman, Paul, MS139, 9:10 Mon
Bhowmick, Sanjukta, MS100, 9:10 Sun
Borggaard, Jeff, MS267, 10:55 Wed
Bauman, Paul, MS139, 9:10 Mon
Bhowmick, Sanjukta, MS125, 1:30 Sun
Börm, Steffen, MS203, 10:00 Tue
Bauman, Paul, MS165, 1:30 Mon
Bhowmick, Sanjukta, MS125, 2:45 Sun
Boscarino, Sebastiano, MS188, 10:50 Tue
Bayona, Victor, MS24, 11:05 Sat
Bian, Xin, PP104, 4:30 Sun
Bosch, Jessica, MS83, 9:35 Sun
Beams, Natalie N., PP6, 4:30 Sun
Bienstock, Daniel, MS100, 10:00 Sun
Bosilca, George, MS82, 9:35 Sun
Beattie, Christopher A., MS91, 9:10 Sun
Biglari, Amir, CP17, 10:10 Wed
Bosilca, George, MS145, 10:00 Mon
Beattie, Christopher A., MS116, 1:30 Sun
Bigoni, Daniele, PP5, 4:30 Sun
Bosse, Torsten F., MS69, 4:35 Sat
Beattie, Christopher A., MS143, 9:10 Mon
Bihari, Barna, MS97, 10:00 Sun
Bourrier, Anthony, MS276, 11:45 Wed
Beattie, Christopher A., MS169, 1:30 Mon
Bilionis, Ilias, MS289, 2:00 Wed
Boutsidis, Christos, MS266, 10:55 Wed
Beck, Andrea D., MS67, 4:35 Sat
Bindel, David, MS60, 4:35 Sat
Beck, Andrea D., MS150, 10:00 Mon
Bindel, David, MS60, 5:00 Sat
Beck, James, MS132, 9:10 Mon
Birken, Philipp, MS6, 10:40 Sat
Beck, James, MS158, 1:30 Mon
Birken, Philipp, MS57, 4:35 Sat
Becker, Stephen, MS35, 2:50 Sat
Biros, George, MS8, 10:15 Sat
Becker, Stephen, MS74, 4:35 Sat
Biros, George, MS34, 2:25 Sat
Beckvermit, Jacqueline, PP206, 4:30 Mon
Biros, George, MS27, 2:50 Sat
Bekas, Costas, MS118, 2:45 Sun
Biros, George, MS59, 4:35 Sat
Belanger-Rioux, Rosalie, MS88, 10:00 Sun
Biros, George, MS164, 2:45 Mon
Bell, John B., MS263, 11:20 Wed
Biros, George, MS214, 2:15 Tue
Bencomo, Mario, MS88, 9:35 Sun
Biros, George, MS239, 4:25 Tue
Bennett, Janine C., MS129, 9:35 Mon
Briggs, Nathan C., PP7, 4:30 Sun
Birrell, Jeremiah, PP6, 4:30 Sun
Benson, Austin, CP4, 9:25 Wed
Brodtkorb, Andre R., MS294, 2:25 Wed
Bistrian, Diana, MS267, 12:10 Wed
Benson, Thomas, CP2, 9:40 Wed
Brown, Elisabeth M., PP4, 4:30 Sun
Bittner, Kai, MS162, 1:55 Mon
Benzi, Michele, MS108, 2:45 Sun
Brown, Jed, MS175, 2:45 Mon
Blackburn, Hugh, MS103, 10:00 Sun
Berger, Marsha, MS130, 9:10 Mon
Blackwell, Robert, MS122, 2:45 Sun
Berger, Marsha, MS157, 1:30 Mon
Blais, Marcel, MS222, 2:15 Tue
Berljafa, Mario, PP3, 4:30 Sun
Blaylock, Myra, MS135, 10:00 Mon
Bernholdt, David E., MS112, 1:55 Sun
Bliss, Nadya, MS100, 10:25 Sun
Bertozzi, Andrea L., MS29, 3:15 Sat
Blom, David, MS6, 11:30 Sat
Bertozzi, Andrea L., PD0, 8:30 Sat
Blonigan, Patrick, MS111, 2:20 Sun Blum, Volker, MS184, 10:00 Tue Bo, Wurigen, MS167, 4:25 Tue
Boyd, John P., MS180, 1:30 Mon Brady, John F., MS288, 2:00 Wed Bramas, Berenger, MS177, 2:20 Mon Brannick, James, MS254, 4:25 Tue Brannick, James, MS254, 4:50 Tue Braun, Richard, PD0, 6:30 Sat Braun, Richard, PP4, 4:30 Sun Brazell, Michael, MS176, 2:45 Mon Breil, Jerome, MS167, 5:15 Tue Brennan, Brian W., MS231, 2:40 Tue Breuer, Alexander, CP8, 9:55 Wed
Brown, Jed, MS202, 10:00 Tue Brown, Jed, MS226, 2:15 Tue Brown, Peter, PP105, 4:30 Sun Brueck, Sascha, MS297, 2:00 Wed Brull, Stephane, MS95, 9:10 Sun Brull, Stephane, CP12, 10:10 Wed Brune, Christoph, MS239, 4:50 Tue Bruno, Oscar P., MS18, 10:15 Sat Bruno, Oscar P., MS154, 9:10 Mon
184
2015 SIAM Conference on Computational Science and Engineering
Brunton, Steven, MS2, 10:15 Sat
Calderer, Antoni E., MS196, 11:15 Tue
Cerioni, Alessandro, MS273, 12:10 Wed
Brunton, Steven, MS28, 2:25 Sat
Calderer, Carme, MS14, 10:40 Sat
Chabannes, Vincent, PP204, 4:30 Mon
Brunton, Steven, MS53, 4:35 Sat
Calhoun, Donna, MS102, 9:10 Sun
Chacon, Luis, MS295, 2:25 Wed
Brunton, Steven, MS53, 4:35 Sat
Calhoun, Donna, MS102, 9:10 Sun
Chakrabortty, Aranya, MS256, 11:45 Wed
Bryant, Corey M., MS109, 2:45 Sun
Calhoun, Donna, MS127, 1:30 Sun
Chakraborty, Pritam, MS259, 11:20 Wed
Bu, Yiming, MS275, 12:10 Wed
Calhoun, Donna, PP102, 4:30 Sun
Challacombe, Matt, MS241, 4:50 Tue
Buchan, Andrew G., MS292, 3:15 Wed
Calvetti, Daniela, MS5, 10:40 Sat
Chand, Kyle, MS127, 2:45 Sun
Bui-Thanh, Tan, MS62, 4:35 Sat
Calvetti, Daniela, PD5, 12:45 Tue
Chandrasekaran, Sunita, MS235, 4:25 Tue
Bui-Thanh, Tan, MS87, 9:10 Sun
Calvin, Christophe, MS17, 11:05 Sat
Chapman, Barbara, MS235, 4:25 Tue
Bui-Thanh, Tan, MS272, 10:55 Wed
Camier, Jean-Sylvain, MS279, 11:45 Wed
Chatzi, Eleni, MS158, 1:30 Mon
Campbell, Dave A., MS158, 2:20 Mon
Chaudhri, Anuj, MS263, 11:45 Wed
Canic, Suncica, MS57, 4:35 Sat
Chaudhry, Jehanzeb H., MS137, 10:25 Mon
Canning, Andrew M., PP3, 4:30 Sun
Chaudhry, Jehanzeb H., MS261, 12:10 Wed
Cantin, Pierre, PP202, 4:30 Mon
Chavarria, Daniel, MS235, 5:40 Tue
Cantwell, Chris, MS26, 10:15 Sat
Chavez, Gustavo, MS73, 5:25 Sat
Cantwell, Chris, MS51, 2:25 Sat
Chelikowsky, James R., MS273, 10:55 Wed
Cantwell, Chris, MS51, 3:15 Sat
Chen, Guangye, MS199, 10:25 Tue
Cantwell, Chris, MS103, 9:10 Sun
Chen, Jie, MS241, 4:25 Tue
Cantwell, Chris, MS103, 9:10 Sun
Chen, Jie, MS241, 4:25 Tue
Cantwell, Chris, MS128, 1:30 Sun
Chen, Long-qing, MS40, 3:40 Sat
Cao, Shuhao, MS152, 10:00 Mon
Chen, Peng, MS34, 3:40 Sat
Carey, Varis, MS84, 10:25 Sun
Chen, Peng, MS187, 10:50 Tue
Carin, Larry, MS276, 10:55 Wed
Chen, Qingshan, PP9, 4:30 Mon
Carlberg, Kevin T., MS4, 10:40 Sat
Chen, Qingshan, MS268, 10:55 Wed
Carlberg, Kevin T., MS187, 10:00 Tue
Chen, Qingshan, MS293, 2:00 Wed
Carlberg, Kevin T., MS211, 2:15 Tue
Chen, Qingshan, MS293, 2:50 Wed
Carlberg, Kevin T., MS236, 4:25 Tue
Chen, Richard L., MS195, 10:25 Tue
Bui-Thanh, Tan, MS296, 2:00 Wed Bui-Thanh, Tan, MS296, 2:00 Wed Bungartz, Hans-Joachim, MS80, 9:10 Sun Bungartz, Hans-Joachim, MS105, 1:30 Sun Burke, James V., MS9, 10:15 Sat Burkovska, Olena, PP2, 4:30 Sun Burkow, Markus, CP24, 9:10 Wed Burstedde, Carsten, MS102, 9:10 Sun Burstedde, Carsten, MS127, 1:30 Sun Burstedde, Carsten, MS127, 1:30 Sun Busetto, AlbertoGiovanni, MS132, 9:10 Mon Busetto, AlbertoGiovanni, MS132, 9:10 Mon Busetto, AlbertoGiovanni, MS158, 1:30 Mon Busetto, AlbertoGiovanni, PP208, 4:30 Mon Butler, Troy, MS84, 9:10 Sun Butler, Troy, MS84, 9:10 Sun Butler, Troy, MS109, 1:30 Sun Butler, Troy, PP101, 4:30 Sun Buvoli, Tommaso, PP10, 4:30 Mon
Carlberg, Kevin T., MS260, 10:55 Wed Carlberg, Kevin T., MS285, 2:00 Wed Carpentieri, Bruno, MS275, 10:55 Wed
C
Cabal, Antonio, MS92, 10:00 Sun Cacuci, Dan G., MS194, 10:25 Tue Cai, Yongyong, CP20, 9:10 Wed Cai, Zhenning, PP105, 4:30 Sun Cai, Zhenning, MS248, 4:25 Tue
Carrola, John, MS16, 10:15 Sat Carrola, John, MS42, 2:25 Sat Carrola, John, MS42, 2:25 Sat Carson, Erin C., MS141, 10:00 Mon Carver, Jeffrey C., MS112, 2:45 Sun Casadei, Astrid, MS250, 5:15 Tue
Chen, Yanping, PP6, 4:30 Sun Chen, Ye, MS299, 3:15 Wed Chen, Yi, CP22, 9:10 Wed Chen, Zheng, MS89, 9:10 Sun Cheng, Cheng, PP8, 4:30 Sun Cheng, Juan, MS23, 10:15 Sat Cheng, Juan, MS23, 10:15 Sat Cheng, Juan, MS49, 2:25 Sat Cheng, Yingda, MS15, 10:15 Sat
Casas, Marc, MS107, 1:30 Sun
Cheng, Yingda, MS64, 4:35 Sat
Catalyurek, Umit V., MS250, 4:50 Tue
Cheng, Yingda, MS70, 5:50 Sat Cheng, Yuanzhen, PP2, 4:30 Sun
Cai, Zhiqiang, MS178, 1:30 Mon
Caudillo Mata, Luz Angelica A., CP9, 9:40 Wed
Calaf, Marc, MS253, 5:15 Tue
Cazeaux, Paul, CP12, 9:25 Wed
Chi, Hongmei, PP13, 4:30 Mon
Cai, Zhiqiang, MS152, 9:10 Mon Cai, Zhiqiang, MS152, 9:10 Mon
Calderbank, Robert, MS300, 2:25 Wed
Chevalier, Cédric, MS200, 10:25 Tue
2015 SIAM Conference on Computational Science and Engineering
185
Chien, Andrew A., MS82, 10:00 Sun
Constantinescu, Emil M., MS188, 10:00 Tue
Davis, Anthony B., PP105, 4:30 Sun
Childs, Hank, PD2, 12:15 Sun
Constantinescu, Emil M., MS212, 2:15 Tue
Davis, Brisa, PP102, 4:30 Sun
Chkifa, Abdellah, MS106, 1:55 Sun
Constantinescu, Emil M., MS237, 4:25 Tue
Davis, Rachel A., MS198, 11:15 Tue
Cho, Hana, PP2, 4:30 Sun
Cook, Dennis, MS7, 10:40 Sat
Davis, Timothy A., MS163, 1:30 Mon
Cho, Heyrim, MS289, 2:50 Wed
Cooley, Kilian, MS228, 2:40 Tue
Dawson, Clint, MS206, 10:00 Tue
Choi, Youngsoo, MS148, 9:35 Mon
Copos, Calina A., PP4, 4:30 Sun
Dawson, Clint, MS206, 10:00 Tue
Chou, Ching-Shan, MS277, 12:10 Wed
Cortez, Ricardo, MS99, 9:10 Sun
Dawson, Clint, MS230, 2:15 Tue
Chow, Edmond, MS197, 11:15 Tue
Cortez, Ricardo, MS140, 9:10 Mon
D’Azevedo, Eduardo F., CP4, 9:10 Wed
Chow, Joe, MS281, 3:15 Wed
Cortial, Julien, MS285, 2:00 Wed
de Jong, Bert, MS209, 3:05 Tue
Christlieb, Andrew J., MS45, 2:50 Sat
Cotilla-Sanchez, Eduardo, MS281, 2:50 Wed
De Marchi, Stefano, MS101, 10:00 Sun
Christlieb, Andrew J., MS136, 9:35 Mon
Cottam, Joseph, MT1, 2:50 Wed
De Stefano, Michele, MS181, 2:20 Mon
Christlieb, Andrew J., MS204, 10:00 Tue
Cotter, Colin J., MS296, 2:50 Wed
De Sterck, Hans, MS131, 9:10 Mon
Christlieb, Andrew J., MS228, 2:15 Tue
Couteyen Carpaye, Jean Marie, PP1, 4:30 Sun
De Sterck, Hans, MS247, 4:25 Tue
Christlieb, Andrew J., MS251, 4:25 Tue Christon, Mark, MS77, 5:25 Sat Chunfeng, Cui, PP3, 4:30 Sun Chung, Julianne, MS31, 2:25 Sat Cinnella, Paola, MS67, 5:00 Sat Clayton, Richard, MS26, 10:15 Sat Clayton, Richard, MS51, 2:25 Sat Cline, Derek A., PP1, 4:30 Sun Cohen, Albert, MS81, 9:10 Sun Cohen, Albert, MS106, 1:30 Sun Cohen, Albert, MS133, 9:10 Mon Cohen, Albert, MS133, 10:00 Mon Cohen, Albert, MS159, 1:30 Mon Cohen, Albert, MS185, 10:00 Tue Coifman, Ronald, MS53, 5:00 Sat
Crabtree, George, MS14, 11:05 Sat Crestel, Benjamin, PP13, 4:30 Mon Crivelli, Silvia N., MS198, 10:00 Tue Crivelli, Silvia N., MS198, 10:00 Tue Crowe, Cameron, PP8, 4:30 Sun
De Sturler, Eric, MS91, 9:10 Sun De Sturler, Eric, MS116, 1:30 Sun De Sturler, Eric, MS116, 2:20 Sun De Sturler, Eric, MS143, 9:10 Mon De Sturler, Eric, MS169, 1:30 Mon
Cui, Tiangang, MS59, 5:00 Sat
Debusschere, Bert J., MS138, 9:10 Mon
Curtis, Frank E., MS244, 5:15 Tue
Debusschere, Bert J., MS164, 1:30 Mon
Cyr, Eric C., PP106, 4:30 Sun
Del Pino, Stephane, MS167, 5:40 Tue Del Razo, Mauricio J., PP102, 4:30 Sun
D Daescu, Dacian N., MS168, 1:30 Mon Daescu, Dacian N., MS194, 10:00 Tue Daescu, Dacian N., MS194, 10:00 Tue Daescu, Dacian N., MS218, 2:15 Tue
Del Razo, Mauricio J., CP9, 9:55 Wed Delgado, Paul M., MS119, 2:02 Sun Delgado, Paul M., PP6, 4:30 Sun D’Elia, Marta, PP201, 4:30 Mon Delong, Steven D., MS263, 12:10 Wed Demanet, Laurent, MS74, 5:50 Sat
Coletti, Mark, PP11, 4:30 Mon
Daescu, Dacian N., MS243, 4:25 Tue
DeMarco, Christopher, MS256, 11:20 Wed
Colgin, Zane, CP13, 9:55 Wed
Daescu, Dacian N., MS267, 10:55 Wed
Demeshko, Irina, MS43, 3:40 Sat
Collins, James B., PP5, 4:30 Sun
Daescu, Dacian N., MS292, 2:00 Wed
Demeshko, Irina, PP106, 4:30 Sun
Conrad, Patrick R., MS74, 5:00 Sat
Dahm, Johann, MS87, 10:00 Sun
Demirci, Utkan, MS42, 2:50 Sat
Conrad, Patrick R., PP5, 4:30 Sun
Dahmen, Wolfgang, IP2, 1:30 Sat
Demlow, Alan, MS178, 1:30 Mon
Constantine, Paul, MS7, 10:15 Sat
Dahmen, Wolfgang, MS81, 10:00 Sun
Dener, Alp, MS205, 10:25 Tue
Constantine, Paul, MS33, 2:25 Sat
Dalton, Steven, PP205, 4:30 Mon
Deng, Weishan, CP10, 9:55 Wed
Darve, Eric F., MS145, 10:25 Mon
Deng, Xiaodi, MS298, 3:15 Wed
Dastouri, Zahrasadat, PP1, 4:30 Sun
Deng, Xiao-Long, MS196, 10:25 Tue
Daunizeau, Jean, CP26, 9:40 Wed
Densmore, Jeffery D., MS223, 2:15 Tue
Davidovic, Andjela, PP12, 4:30 Mon
Derksen, Alexander, MS214, 2:40 Tue
Davis, Anthony B., MS19, 10:15 Sat
Desjardins, Olivier, MS52, 5:50 Sat
Constantine, Paul, MS58, 4:35 Sat Constantine, Paul, MS249, 4:25 Tue Constantinescu, Emil M., MS136, 9:10 Mon Constantinescu, Emil M., MS162, 1:30 Mon Constantinescu, Emil M., MS188, 10:00 Tue
186
2015 SIAM Conference on Computational Science and Engineering
Deslippe, Jack, MS234, 5:40 Tue
Doostan, Alireza, MS289, 2:00 Wed
Endeve, Eirik, MS95, 9:35 Sun
Despres, Bruno, MS48, 2:25 Sat
Dorff, Michael, MS222, 3:30 Tue
Engsig-Karup, Allan P., PP10, 4:30 Mon
Detrixhe, Miles L., CP11, 10:10 Wed
Dow, Eric, MS7, 10:15 Sat
Engwer, Christian, MS231, 3:30 Tue
Deveci, Mehmet, MS287, 2:25 Wed
Draganescu, Andrei, MS202, 10:25 Tue
Ennis-King, Jonathan, MS284, 2:50 Wed
Devendran, Dharshi, MS157, 1:30 Mon
Drawert, Brian, MS90, 9:10 Sun
Enright, Wayne, MS188, 10:25 Tue
Devine, Karen D., MS200, 10:00 Tue
Drawert, Brian, MS115, 1:30 Sun
Epshteyn, Yekaterina, MS10, 10:15 Sat
Devine, Karen D., MS200, 10:50 Tue
Driscoll, Tobin, MS180, 2:45 Mon
Epshteyn, Yekaterina, MS13, 10:15 Sat
Devine, Karen D., MS250, 4:25 Tue
Drmac, Zlatko, MS169, 1:30 Mon
Epshteyn, Yekaterina, MS36, 2:25 Sat
DeVore, Ronald, MS81, 9:10 Sun
Druinsky, Alex, MS210, 3:05 Tue
Epshteyn, Yekaterina, MS39, 2:25 Sat
Dexter, Nick, MS240, 5:40 Tue
Druskin, Vladimir L., MS143, 9:10 Mon
Epshteyn, Yekaterina, MS251, 5:15 Tue
Di, Zichao, MS202, 10:00 Tue
Du, Cheng-Han, CP10, 10:10 Wed
Español, Pep, MS263, 10:55 Wed
Di, Zichao, MS202, 10:00 Tue
Duarte, Marco F., MS300, 3:15 Wed
Espig, Mike, MS56, 4:35 Sat
Di, Zichao, MS226, 2:15 Tue
Duarte, Max, CP10, 9:25 Wed
Espig, Mike, MS56, 4:35 Sat
Duarte, Max, CP17, 9:55 Wed
Estrada, Felipe, MS216, 2:40 Tue
Dubey, Anshu, MS160, 2:20 Mon
Evans, Emily, CP19, 9:40 Wed
Di Napoli, Edoardo A., MS273, 10:55 Wed Di Napoli, Edoardo A., MS297, 2:00 Wed Diachin, Lori A., MS201, 10:00 Tue Diachin, Lori A., MS225, 2:15 Tue Diaz, Julien, MS277, 11:20 Wed Dietrich, J. Casey, MS230, 3:05 Tue Dimarco, Giacomo, MS173, 2:45 Mon DiStasio, Jr., Robert A., MS184, 10:25 Tue Ditkowsky, Adi, MS10, 10:40 Sat Dobson, Matthew, PP6, 4:30 Sun Doelz, Juergen, MS159, 2:45 Mon Dogan, Gunay, PP3, 4:30 Sun Dogan, Gunay, MS226, 2:15 Tue Dogdas, Belma, MS117, 1:30 Sun Domingues, Margarete O., CP23, 10:10 Wed Donev, Aleksandar, MS193, 10:50 Tue Donev, Aleksandar, MS263, 10:55 Wed Donev, Aleksandar, MS288, 2:00 Wed Donev, Aleksandar, MS288, 3:15 Wed Dong, Bo, MS23, 10:40 Sat Dong, Suchuan, CP17, 9:10 Wed Dongarra, Jack J., MS266, 10:55 Wed Dongarra, Jack J., MS291, 2:00 Wed Doostan, Alireza, MS191, 10:00 Tue Doostan, Alireza, MS215, 2:15 Tue Doostan, Alireza, MS240, 4:25 Tue Doostan, Alireza, MS264, 10:55 Wed Doostan, Alireza, MS264, 10:55 Wed
Dutta, Prashanta, MS217, 3:30 Tue Dwight, Richard, MS135, 9:35 Mon Dytrych, Tomas, MS61, 5:25 Sat
E
Ebna Hai, Bhuiyan Shameem M., PP10, 4:30 Mon Economon, Thomas, MS302, 3:15 Wed Edelman, Alan, MS246, 4:25 Tue Edwards, John, PP4, 4:30 Sun Efendiev, Yalchin, MS290, 2:00 Wed Ehrlacher, Virginie, MS224, 3:30 Tue Eldred, Michael S., MS215, 3:30 Tue Ellingson, Sally R., PP6, 4:30 Sun Ellingwood, Nathan, PP4, 4:30 Sun Elliott, James, MS134, 10:25 Mon Elvetun, Ole Løseth, MS252, 4:50 Tue Elwasif, Wael R., MS171, 2:20 Mon Emad, Nahid, MS68, 5:50 Sat Embree, Mark, MS108, 1:55 Sun Emelianenko, Maria, PD0, 6:30 Sat
F
Fabiano, Enrico, MS86, 9:35 Sun Fahroo, Fariba, PD0, 8:30 Sat Fairbanks, James, PP3, 4:30 Sun Falgout, Robert, IP4, 11:20 Sun Falk, Michael, MS40, 3:15 Sat Fan, Yuwei, PP105, 4:30 Sun Fang, Fangxin, MS267, 11:20 Wed Fange, David, MS90, 9:35 Sun Farge, Marie, PP1, 4:30 Sun Farrell, Kathryn, PP5, 4:30 Sun Farrell, Patricio, MS75, 5:00 Sat Farrell, Patrick E., MS182, 10:25 Tue Farrell, Patrick E., MS292, 2:50 Wed Fattebert, Jean-Luc, MS209, 2:15 Tue Feinberg, Eugene A., CP15, 9:25 Wed Feng, Lihong, MS143, 10:00 Mon Feng, Xiao, MS251, 5:40 Tue Ferreira, Ricardo, MS198, 10:25 Tue Ferrer, Rodolfo, MS173, 1:55 Mon
Emelianenko, Maria, PD0, 8:30 Sat
Fidkowski, Krzysztof, MS98, 10:25 Sun
Emelianenko, Maria, MS202, 10:00 Tue
Field, Scott, PP10, 4:30 Mon
Emelianenko, Maria, MS226, 2:15 Tue
Field, Scott, MS301, 2:25 Wed
Emerson, David B., MS137, 9:35 Mon
Fike, Jeffrey, MS211, 2:40 Tue
Emory, Michael A., MS58, 5:25 Sat
Filbir, Frank, MS126, 1:30 Sun
2015 SIAM Conference on Computational Science and Engineering
187
G
Ghosh, Debojyoti, MS228, 3:05 Tue
Gadou, Mohamed, MS163, 2:20 Mon
Ghysels, Pieter, MS266, 11:20 Wed
Flyer, Natasha, MS24, 10:15 Sat
Galagusz, Ryan, MS66, 5:25 Sat
Gibou, Frederic G., MS196, 10:00 Tue
Fogelson, Aaron L., MS16, 10:40 Sat
Galanti, Barak, PP102, 4:30 Sun
Gibou, Frederic G., MS196, 10:00 Tue
Fornasier, Massimo, MS7, 11:30 Sat
Galaz, José, MS119, 2:18 Sun
Gibou, Frederic G., MS220, 2:15 Tue
Forstall, Virginia, MS260, 11:45 Wed
Galvin-Donoghue, Mary, MS65, 5:50 Sat
Gibou, Frederic G., MS245, 4:25 Tue
Fountoulakis, Kimon, MS179, 2:20 Mon
Gambino, James R., PP1, 4:30 Sun
Gie, Gung-Min, MS268, 10:55 Wed
Fountoulakis, Vasileios, PP15, 4:30 Mon
Gamell, Marc, MS107, 1:55 Sun
Gie, Gung-Min, MS268, 11:45 Wed
Fox, Alyson, CP2, 10:10 Wed
Gandham, Rajesh, MS294, 2:50 Wed
Fox, Rodney O., MS19, 10:40 Sat
Gie, Gung-Min, MS293, 2:00 Wed
Ganesh, Mahadevan, CP16, 9:40 Wed
Franchetti, Franz, MS283, 2:25 Wed
Giffen, Deena H., CP21, 9:55 Wed
Ganguli, Surya, MS53, 5:25 Sat
Franck, Isabell, CP5, 10:10 Wed
Giles, Michael B., MS133, 9:10 Mon
Gansterer, Wilfried N., MS160, 2:45 Mon
Francois, Marianne M., MS52, 5:00 Sat
Gillespie, Dirk, MS153, 10:00 Mon
Garcke, Jochen, MS185, 10:00 Tue
Gillette, Andrew, PP207, 4:30 Mon
Frank, Martin, MS19, 10:15 Sat
Gardner, Carl L., MS153, 9:10 Mon
Gillman, Adrianna, MS48, 2:25 Sat
Frank, Martin, MS45, 2:25 Sat
Gardner, Carl L., MS153, 9:35 Mon
Gillman, Adrianna, MS73, 4:35 Sat
Frank, Martin, MS70, 4:35 Sat
Gardner, David J., MS201, 10:25 Tue
Gillman, Adrianna, MS203, 11:15 Tue
Frank, Martin, MS95, 9:10 Sun
Garg, Vikram, MS34, 3:15 Sat
Gilyard, Julian, MS221, 3:05 Tue
Frank, Martin, MS120, 1:30 Sun
Garrett, Charles K., PP105, 4:30 Sun
Frank, Martin, PP105, 4:30 Sun
Garrett, Charles K., MS199, 10:00 Tue
Frank, Martin, MS147, 9:10 Mon
Gaston, Derek R., MS182, 10:00 Tue
Frank, Martin, MS173, 1:30 Mon
Gaston, Derek R., MS259, 10:55 Wed
Frank, Martin, MS199, 10:00 Tue
Gaston, Derek R., MS284, 2:00 Wed
Frank, Martin, MS223, 2:15 Tue
Gauger, Nicolas R., MS69, 4:35 Sat
Franzelin, Fabian, CP16, 9:10 Wed
Gauger, Nicolas R., MS86, 9:10 Sun
Frazier, Peter I., MS183, 10:00 Tue
Gauger, Nicolas R., MS86, 10:00 Sun
Frean, Daniel, MS228, 3:30 Tue
Gauger, Nicolas R., MS111, 1:30 Sun
Gleich, David F., MS1, 10:15 Sat
Friedhoff, Stephanie, MS261, 10:55 Wed
Gauger, Nicolas R., MS302, 2:00 Wed
Glimm, James G., MS233, 4:25 Tue
Friedlander, Michael, MS9, 10:40 Sat
Gawlik, Evan S., CP9, 9:25 Wed
Gluck, Fred, MS232, 5:15 Tue
Fringer, Oliver, MS230, 2:40 Tue
Gazzola, Silvia, MS31, 2:50 Sat
Gmeiner, Björn, MS221, 2:15 Tue
Frischknecht, Amalie, MS217, 3:05 Tue
Gelb, Anne, MS154, 9:35 Mon
Go, Jaegwi, PP4, 4:30 Sun
Froese, Brittany, MS64, 5:50 Sat
Gentile, Ann, MS262, 11:20 Wed
Gobbert, Matthias K., MS197, 10:00 Tue
Fromm, Bradley, MS259, 10:55 Wed
Ghanem, Roger, MS240, 4:25 Tue
Fu, Guosheng, MS176, 1:55 Mon
Ghattas, Omar, MS8, 10:15 Sat
Fukaya, Takeshi, MS63, 5:25 Sat
Godinez, Humberto C., MS243, 5:40 Tue
Ghattas, Omar, MS46, 2:25 Sat
Funke, Simon W., PP204, 4:30 Mon
Goldberg, Daniel, MS190, 10:25 Tue
Ghattas, Omar, MS34, 2:25 Sat
Gonnet, Pedro, MS270, 11:20 Wed
Funke, Simon W., MS229, 2:15 Tue
Ghattas, Omar, MS71, 4:35 Sat
Goodfriend, Lauren, MS102, 10:00 Sun
Funke, Simon W., MS231, 2:15 Tue
Ghattas, Omar, MS59, 4:35 Sat
Gorman, Gerard J, MS128, 1:55 Sun
Funke, Simon W., MS252, 4:25 Tue
Ghattas, Omar, PD0, 8:30 Sat
Graf, Jonathan, MS197, 10:00 Tue
Fuselier, Edward, MS101, 10:25 Sun
Ghattas, Omar, MS116, 1:30 Sun
Graham, Erica J., MS113, 1:30 Sun
Ghesmati, Arezou, PP8, 4:30 Sun
Graham, Erica J., MS113, 1:30 Sun
Fischer, Paul F., MS128, 1:30 Sun Flenner, Arjuna, MS3, 10:40 Sat
Gholaminejad, Amir, MS239, 5:40 Tue
Giraud, Luc, MS82, 9:10 Sun Giraud, Luc, MS107, 1:30 Sun Giraud, Luc, MS134, 9:10 Mon Giraud, Luc, MS160, 1:30 Mon Girolami, Mark, MS132, 10:00 Mon Glas, Silke, MS285, 3:15 Wed Glawe, Christoph, MS20, 11:30 Sat Gleich, David F., MS1, 10:15 Sat
Gobbert, Matthias K., MS221, 2:15 Tue
188
2015 SIAM Conference on Computational Science and Engineering
Graham, Lindley C., MS84, 9:35 Sun
Guittet, Arthur, CP17, 9:40 Wed
Graham, Lindley C., PP101, 4:30 Sun
Günther, Stefanie, MS69, 5:25 Sat
Gramacy, Robert, MS183, 10:25 Tue
Gunzburger, Max, MS81, 9:10 Sun
Grandine, Thomas A., PD0, 6:30 Sat
Gunzburger, Max, MS106, 1:30 Sun
Grandine, Tom, PD1, 12:15 Sat
Gunzburger, Max, MS133, 9:10 Mon
Grant, Zachary J., PP10, 4:30 Mon
Halappanavar, Mahantesh, MS281, 2:00 Wed
Gunzburger, Max, MS159, 1:30 Mon
Grant, Zachary J., MS212, 3:30 Tue
Hall, David M., MS272, 11:45 Wed
Gunzburger, Max, PP201, 4:30 Mon
Granzow, Brian, PP103, 4:30 Sun
Hall, Mary, MS258, 11:45 Wed
Gunzburger, Max, MS185, 10:00 Tue
Granzow, Brian, MS225, 3:05 Tue
Ham, David, PP204, 4:30 Mon
Guo, Hanliang, MS22, 11:30 Sat
Grasedyck, Lars, MS56, 5:00 Sat
Ham, David, MS207, 10:00 Tue
Guo, Hong, CP11, 9:25 Wed
Gratadour, Damien, MS93, 10:00 Sun
Ham, David, MS207, 10:00 Tue
Guo, Wei, PP105, 4:30 Sun
Graziani, Frank, MS147, 9:10 Mon
Ham, David, MS231, 2:15 Tue
Guthrey, Pierson, CP12, 9:55 Wed
Greenspan, Elizabeth, PD5, 12:45 Tue
Hamlet, Christina, MS16, 11:30 Sat
Guzzetti, Sofia, CP21, 10:10 Wed
Hammett, Greg, PP105, 4:30 Sun
Greif, Chen, MS83, 10:00 Sun Griewank, Andreas, MS194, 10:50 Tue Griffith, Boyce, MS16, 11:05 Sat
H
Halappanavar, Mahantesh, MS256, 10:55 Wed Halappanavar, Mahantesh, MS269, 12:10 Wed
Hampton, Jerrad, MS238, 4:25 Tue Hangelbroek, Thomas C., MS101, 9:35 Sun
Höft, Thomas, PP6, 4:30 Sun
Hansen, Anders, MS276, 10:55 Wed
Haack, Jeffrey, MS120, 1:55 Sun
Hansen, Anders, MS300, 2:00 Wed
Haario, Heikki, MS59, 5:50 Sat
Hansen, Glen, MS304, 2:00 Wed
Haasdonk, Bernard, MS4, 10:15 Sat
Hansen, Michael A., PP9, 4:30 Mon
Haasdonk, Bernard, MS148, 9:10 Mon
Harmon, Michael D., PP7, 4:30 Sun
Haasdonk, Bernard, MS174, 1:30 Mon
Hasler, Jennifer, MS153, 10:25 Mon
Grove, John W., MS208, 2:40 Tue
Haber, Eldad, MS91, 9:10 Sun
Hauck, Cory, MS19, 10:15 Sat
Grove, Ryan R., PP2, 4:30 Sun
Haber, Eldad, MS91, 9:10 Sun
Hauck, Cory, MS45, 2:25 Sat
Gu, Li, PP201, 4:30 Mon
Haber, Eldad, MS116, 1:30 Sun
Hauck, Cory, MS70, 4:35 Sat
Gu, Tongxiang, CP7, 9:40 Wed
Haber, Eldad, MS143, 9:10 Mon
Hauck, Cory, MS95, 9:10 Sun
Gu, Xiaojun, MS248, 4:25 Tue
Haber, Eldad, MS169, 1:30 Mon
Hauck, Cory, MS120, 1:30 Sun
Guarin Zapata, Nicolas, PP4, 4:30 Sun
Haber, Eldad, PD5, 12:45 Tue
Hauck, Cory, MS147, 9:10 Mon
Guenther, Michael, MS6, 10:15 Sat
Haber, Eldad, MS246, 4:25 Tue
Hauck, Cory, MS173, 1:30 Mon
Guenther, Michael, MS136, 9:10 Mon
Haberl, Alexander, MS227, 3:05 Tue
Hauck, Cory, MS199, 10:00 Tue
Guenther, Michael, MS162, 1:30 Mon
Hadri, Bilel, MS93, 9:10 Sun
Hauck, Cory, MS223, 2:15 Tue
Guenther, Michael, MS188, 10:00 Tue
Hadri, Bilel, MS118, 1:30 Sun
Hawkins-Daarud, Andrea, MS166, 1:55 Mon
Guenther, Michael, MS212, 2:15 Tue
Hagstrom, Thomas M., MS277, 10:55 Wed
Hawkins-Daarud, Andrea, MS239, 5:15 Tue
Guenther, Michael, MS237, 4:25 Tue
Hagstrom, Thomas M., MS277, 10:55 Wed
Haynes, Ronald, MS237, 5:40 Tue
Guerra, Jorge E., MS272, 11:20 Wed
Hagstrom, Thomas M., MS301, 2:00 Wed
Haynes, Ronald, MS261, 10:55 Wed
Gugercin, Serkan, MS91, 9:10 Sun
Haidar, Azzam, MS12, 10:15 Sat
Haynes, Ronald, MS286, 2:00 Wed
Gugercin, Serkan, MS116, 1:30 Sun
Haidar, Azzam, MS12, 10:15 Sat
He, Cuiyu, PP8, 4:30 Sun
Gugercin, Serkan, MS143, 9:10 Mon
Haidar, Azzam, MS38, 2:25 Sat
He, Cuiyu, MS178, 2:20 Mon
Gugercin, Serkan, MS169, 1:30 Mon
Haidar, Azzam, MS63, 4:35 Sat
He, Cuiyu, PP10, 4:30 Mon
Gugercin, Serkan, MS168, 2:20 Mon
Hajghassem, Mona, PP3, 4:30 Sun
He, Xiaoming, MS208, 2:15 Tue
Gugercin, Serkan, MS211, 3:05 Tue
Hakim, Ammar, MS95, 10:00 Sun
He, Xiaoming, MS233, 4:25 Tue
Grigori, Laura, MS141, 9:10 Mon Grimm, Alexander, MS143, 9:35 Mon Grindeanu, Iulian, PP1, 4:30 Sun Gropp, William D., PP204, 4:30 Mon Grotendorst, Johannes, CP3, 10:10 Wed Groth, Clinton P., MS295, 3:15 Wed
2015 SIAM Conference on Computational Science and Engineering
189
He, Xiaoming, MS257, 10:55 Wed
Higgins, Raegan, MS192, 10:00 Tue
Howell, Jason, CP18, 9:10 Wed
He, Xiaoming, MS257, 12:10 Wed
Higgins, Raegan, MS216, 2:15 Tue
Hu, Guanghui, MS204, 11:15 Tue
He, Xiaoming, MS282, 2:00 Wed
Higham, Nicholas J., PD5, 12:45 Tue
Hu, Jingwei, MS45, 3:40 Sat
He, Ying, MS44, 3:15 Sat
Higueras, Inmaculada, MS188, 11:15 Tue
Hu, Jonathan J., PP103, 4:30 Sun
He, Ying, MS89, 9:35 Sun
Hill, Chris, MS190, 11:15 Tue
Hu, Jonathan J., MS149, 9:35 Mon
Hebbur Venkata Subba Rao, Vishwas, MS190, 10:50 Tue
Hillewaert, Koen, MS67, 4:35 Sat
Hu, Jun, MS152, 10:25 Mon
Hillewaert, Koen, MS67, 4:35 Sat
Hu, Jun, PP207, 4:30 Mon
Himpe, Christian, MS116, 1:55 Sun
Hu, Xiaozhe, MS11, 10:40 Sat
Himpe, Christian, PP5, 4:30 Sun
Hu, Zhicheng, MS248, 4:25 Tue
Hittinger, Jeffrey A., PD0, 8:30 Sat
Huang, Chen, MS234, 4:50 Tue
Hittinger, Jeffrey A., MS186, 10:00 Tue
Huang, Chen, CP1, 9:25 Wed
Hittinger, Jeffrey A., MS210, 2:15 Tue
Huang, Jianjun, CP21, 9:40 Wed
Ho, Kenneth L., MS74, 4:35 Sat
Huang, Zhenyu, MS281, 2:00 Wed
Ho, Nguyenho, PP9, 4:30 Mon
Huckle, Thomas K., CP6, 10:10 Wed
Heister, Timo, MS103, 9:35 Sun Heister, Timo, MS155, 9:35 Mon Heister, Timo, PP204, 4:30 Mon Helenbrook, Brian, MS144, 10:00 Mon Hellander, Andreas, MS90, 9:10 Sun Hellander, Andreas, MS115, 1:30 Sun Helzel, Christiane, MS130, 9:10 Mon Henriksen, Ian D., PP6, 4:30 Sun Henshaw, William, MS32, 2:50 Sat Herbert-Voss, Ariel, MS146, 9:26 Mon Heroux, Michael, MS17, 10:15 Sat Heroux, Michael, MS17, 10:15 Sat Heroux, Michael, MS43, 2:25 Sat Heroux, Michael, MS68, 4:35 Sat Heroux, Michael, MS82, 9:10 Sun Heroux, Michael, MS82, 9:10 Sun Heroux, Michael, MS107, 1:30 Sun
Hoefler, Torsten, MS262, 10:55 Wed Hoefler, Torsten, MS287, 2:00 Wed Hoemmen, Mark, MS108, 2:20 Sun Hoffman, Bill, MS77, 4:35 Sat Hoffman, Bill, MS77, 4:35 Sat Hoffman, Johan, PP204, 4:30 Mon Hoffmann, Franz M., MS22, 11:05 Sat Hogg, Jonathan, MS163, 1:55 Mon Hohenegger, Christel, MS96, 9:10 Sun Hohenegger, Christel, MS122, 1:30 Sun
Humpherys, Jeffrey, MS222, 3:05 Tue Humpherys, Jeffrey, MS280, 3:15 Wed Humphrey, Alan, PP206, 4:30 Mon Hunter, Blake, MS3, 11:05 Sat Hutchins, John T., PP2, 4:30 Sun Huybrechs, Daan, MS154, 9:10 Mon Huybrechs, Daan, MS180, 1:30 Mon Hysing, Johan S., CP19, 9:55 Wed
I
Heroux, Michael, MS134, 9:10 Mon
Holgado, Aaron M., PP5, 4:30 Sun
Iaccarino, Gianluca, MS161, 1:30 Mon
Heroux, Michael, MS160, 1:30 Mon
Hollingsworth, Jeffery, MS258, 11:20 Wed
Ibanez, Dan A., MS97, 9:10 Sun
Heroux, Michael, PP204, 4:30 Mon
Holton, Douglas, MS299, 2:50 Wed
Ibanez, Dan A., PP103, 4:30 Sun
Herrmann, Marcus, MS52, 5:25 Sat
Holzaepfel, Aaron, PP105, 4:30 Sun
Ibeid, Huda, MS73, 5:00 Sat
Horesh, Lior, MS74, 4:35 Sat
Ibrahima, Fayadhoi, PP5, 4:30 Sun
Horesh, Lior, MS91, 9:35 Sun
Iglesias, Marco, MS59, 5:25 Sat
Herty, Michael, MS147, 10:25 Mon Heryudono, Alfa, MS75, 5:50 Sat Hess, Martin W., MS169, 2:20 Mon
Horntrop, David J., CP22, 10:10 Wed
Iliescu, Traian, MS211, 2:15 Tue
Heuveline, Vincent, PD2, 12:15 Sun
Horsch, Martin T., CP1, 10:10 Wed
Imakura, Akira, MS134, 9:10 Mon
Hewett, Russell, MS73, 5:50 Sat
Hoshino, Tetsuya, MS43, 2:50 Sat
Imamura, Toshiyuki, MS258, 10:55 Wed
Hewson, John C., MS20, 10:40 Sat
Hötzer, Johannes, MS124, 1:55 Sun
Imamura, Toshiyuki, MS283, 2:00 Wed
Hicken, Jason E., MS111, 1:55 Sun
Hou, Zhangshuan, CP16, 9:25 Wed
Imamura, Toshiyuki, MS283, 2:00 Wed
Hicken, Jason E., MS205, 10:00 Tue
Hovland, Paul D., MS190, 10:00 Tue
Imbert-Gérard, Lise-Marie, MS48, 2:25 Sat
Hickernell, Fred J., MS159, 2:20 Mon
Howard, Marylesa, MS144, 9:35 Mon
Imbert-Gérard, Lise-Marie, MS73, 4:35 Sat
Higgins, Raegan, MS88, 9:10 Sun
Howe, Bill, MS78, 10:00 Sun
Imbert-Gérard, Lise-Marie, MS73, 4:35 Sat
Higgins, Raegan, MS113, 1:30 Sun
Howell, Gary W., CP25, 9:55 Wed
Ipsen, Ilse, MS266, 11:45 Wed
Higgins, Raegan, MS140, 9:10 Mon
Howell, Gary W., MS275, 10:55 Wed
Irish, Jennifer L., MS166, 1:30 Mon
Higgins, Raegan, MS166, 1:30 Mon
190
2015 SIAM Conference on Computational Science and Engineering
Isaac, Toby, MS71, 4:35 Sat
Johansen, Hans, MS296, 2:25 Wed
Karpinski, Stephan, MS182, 10:50 Tue
Iske, Armin, MS75, 5:25 Sat
Johansson, August, MS10, 11:30 Sat
Katagiri, Takahiro, MS258, 10:55 Wed
Iturraran-Viveros, Ursula, PP6, 4:30 Sun
Johnsen, Pete, MS118, 2:20 Sun
Katagiri, Takahiro, MS258, 10:55 Wed
Ivan, Lucian, MS271, 10:55 Wed
Johnson, Calvin W., MS61, 4:35 Sat
Katagiri, Takahiro, MS283, 2:00 Wed
Ivan, Lucian, MS295, 2:00 Wed
Johnson, Christopher, PD4, 12:45 Tue
Katsoulakis, Markos A., PP104, 4:30 Sun
Ivan, Lucian, MS295, 2:00 Wed
Johnson, Christopher, MS247, 4:25 Tue
Katz, Richard F., MS155, 9:10 Mon
Jolivet, Pierre, MS226, 2:40 Tue
Katz, Richard F., MS181, 1:30 Mon
Jordan, Tobias, PP2, 4:30 Sun
Kaus, Boris, MS155, 10:25 Mon
Joseph-Ellison, Stacey, MS94, 9:10 Sun
Keasler, Jeff, MS17, 11:30 Sat
Joseph-Ellison, Stacey, PP7, 4:30 Sun
Keener, James P., MS26, 11:05 Sat
Joshi, Sunnie, MS89, 10:00 Sun
Kees, Chris, CP6, 9:40 Wed
Jouvet, Guillaume, MS46, 3:40 Sat
Kekenes-Huskey, Peter, MS51, 2:50 Sat
Jovanovic, Mihailo R., MS28, 2:50 Sat
Keller, Tobias, MS181, 2:45 Mon
Joyce, Kevin, MS170, 1:55 Mon
Kelley, C.T., PD0, 6:30 Sat
Ju, Lili, MS46, 3:15 Sat
Kelley, C.T., MS142, 9:10 Mon
Juckeland, Guido, MS235, 4:50 Tue
Kelley, C.T., MS142, 9:10 Mon
Jung, Chang-Yeol, MS293, 3:15 Wed
Kennedy, Graeme, MS205, 10:00 Tue
JZhu, Jiang, MS292, 2:00 Wed
Kennedy, Graeme, MS205, 10:00 Tue
J
Jackiewicz, Zdzislaw, MS212, 2:15 Tue Jacquelin, Mathias, MS149, 10:25 Mon Jafarpour, Benham, MS238, 4:50 Tue Jain, Rajeev, MS121, 1:55 Sun Jainta, Marcus, PP6, 4:30 Sun Jakeman, John D., MS189, 10:00 Tue Jakeman, John D., MS213, 2:15 Tue Jakeman, John D., MS213, 2:15 Tue Jakeman, John D., MS238, 4:25 Tue Jalali, Alireza, MS150, 10:25 Mon James, Doug L., MS260, 10:55 Wed James, Richard, MS65, 4:35 Sat
K
Kent, Carson, MS58, 5:00 Sat Kent, Paul, MS209, 3:30 Tue
Kabir, Humayun, MS221, 2:40 Tue
Kenway, Gaetan, MS205, 11:15 Tue
Kågström, Bo T., MS38, 2:25 Sat
Keppens, Rony, MS271, 11:20 Wed
Kahl, Karsten, MS254, 5:15 Tue
Kerfriden, Pierre, MS4, 11:30 Sat
Kaiser, Eurika, MS2, 10:40 Sat
Ketcheson, David I., PP102, 4:30 Sun
Kaizu, Kazunari, MS90, 10:25 Sun
Ketcheson, David I., MS212, 2:40 Tue
Kalashnikova, Irina, MS46, 2:25 Sat
Kevlahan, Nicholas, MS41, 2:25 Sat
Kalashnikova, Irina, MS46, 2:50 Sat
Kevrikidis, Yannis, MS28, 3:40 Sat
Kalashnikova, Irina, MS71, 4:35 Sat
Keyes, David E., PD1, 12:15 Sat
Kallemov, Bakytzhan, MS288, 2:50 Wed
Keyes, David E., MS93, 9:10 Sun
Kandasamy, Manickam, CP14, 10:10 Wed
Keyes, David E., PD3, 12:15 Mon
Kanner, Samuel, MS67, 5:25 Sat
Khabou, Amal, MS266, 12:10 Wed
Kanso, Eva, MS99, 10:25 Sun
Khan, Maryam, MS192, 10:25 Tue
Kaper, Hans G., MS14, 10:15 Sat
Khatri, Shilpa, CP9, 10:10 Wed
Kaper, Hans G., MS40, 2:25 Sat
Khuvis, Samuel, MS197, 10:50 Tue
Jiao, Xiangmin, MS121, 2:45 Sun
Kaper, Hans G., MS65, 4:35 Sat
Kilmer, Misha E., MS5, 10:15 Sat
Jimack, Peter K., CP2, 9:25 Wed
Kara, Rukiye, CP24, 9:55 Wed
Kilmer, Misha E., MS31, 2:25 Sat
Jimenez Bolanos, Silvia, MS113, 1:55 Sun
Karimi, Saeid, PP10, 4:30 Mon
Kilmer, Misha E., MS91, 9:10 Sun
Jin, Shi, MS45, 3:15 Sat
Karimi, Saeid, CP10, 9:40 Wed
Kilmer, Misha E., MS116, 1:30 Sun
Karlsson, Lars, MS38, 2:50 Sat
Kilmer, Misha E., MS143, 9:10 Mon
Karniadakis, George E., PP104, 4:30 Sun
Kilmer, Misha E., MS169, 1:30 Mon
Jameson, Antony, MS98, 9:10 Sun Jameson, Antony, MS123, 1:30 Sun Jameson, Antony, MS123, 1:30 Sun Jameson, Antony, MS150, 9:10 Mon Jameson, Antony, MS176, 1:30 Mon Jandhyala, Vikram, MS54, 5:50 Sat Jansen, Kenneth, PP103, 4:30 Sun Jansen, Kenneth, MS225, 2:15 Tue Jantsch, Peter, MS138, 10:00 Mon Jeannot, Emmanuel, MS262, 12:10 Wed Jefferson, Jennifer, MS58, 4:35 Sat Jiahao, Chen, MS246, 4:25 Tue Jiang, Jiahua, PP8, 4:30 Sun Jiang, Jiahua, PP5, 4:30 Sun
Johansen, Hans, MS112, 1:30 Sun Johansen, Hans, MS130, 9:10 Mon Johansen, Hans, MS157, 1:30 Mon
Karpeyev, Dmitry A., PP204, 4:30 Mon
2015 SIAM Conference on Computational Science and Engineering
191
Kim, Changho, PP6, 4:30 Sun
Kordy, Michal A., PP6, 4:30 Sun
Langtangen, Hans Petter, MS78, 9:35 Sun
Kim, Kyungjoo, MS193, 10:00 Tue
Koumoutsakos, Petros, MS158, 1:55 Mon
Langtangen, Hans Petter, MS182, 10:00 Tue
Kim, Kyungjoo, MS217, 2:15 Tue
Kouri, Drew P., MS244, 5:40 Tue
Lapham, Gary, CP24, 9:40 Wed
Kim, Kyungjoo, MS217, 2:40 Tue
Kouri, Drew P., MS274, 10:55 Wed
Larson, Stephen, MS115, 1:55 Sun
Kim, Kyungjoo, MS242, 4:25 Tue
Kouri, Drew P., MS298, 2:00 Wed
Larsson, Elisabeth, MS75, 4:35 Sat
Kim, Mintae, MS12, 11:05 Sat
Kowalski, Karol, MS234, 5:15 Tue
Lashuk, Ilya, MS137, 10:00 Mon
Kirby, Mike, MS128, 2:20 Sun
Kowitz, Christoph, PP11, 4:30 Mon
Lau, Stephen, PP6, 4:30 Sun
Kirby, Rob, MS207, 10:00 Tue
Kozdon, Jeremy E., MS102, 10:25 Sun
Laurent, Thomas, MS3, 11:30 Sat
Kirby, Rob, MS231, 2:15 Tue
Kramer, Boris, PP11, 4:30 Mon
Law, Kody, MS8, 10:40 Sat
Kirby, Robert C., PP106, 4:30 Sun
Krämer, Lukas, MS297, 2:25 Wed
Law, Kody, MS185, 10:25 Tue
Kitanidis, Peter K., MS34, 2:25 Sat
Kraus, Michael, CP5, 9:40 Wed
Lazar, Emanuel A., CP1, 9:10 Wed
Klatt, Torbjörn, PP14, 4:30 Mon
Krause, Rolf, MS79, 9:10 Sun
Lazarov, Boyan S., CP13, 10:10 Wed
Kloeckner, Andreas, PP106, 4:30 Sun
Kreienbuehl, Andreas, MS261, 11:45 Wed
Le, Ellen B., MS33, 2:25 Sat
Kloeckner, Andreas, MS151, 9:10 Mon
Kreiss, Gunilla, MS10, 10:15 Sat
Kloeckner, Andreas, MS177, 1:30 Mon
Kreiss, Gunilla, MS36, 2:25 Sat
Le-Dimet, Francois-Xavier L., MS194, 11:15 Tue
Kloeckner, Andreas, MS203, 10:00 Tue
Kressner, Daniel, MS224, 2:15 Tue
Kloeckner, Andreas, MS227, 2:15 Tue
Kressner, Daniel, MS249, 4:25 Tue
Kloeckner, Andreas, MS227, 3:30 Tue
Krishnamurthy, Adarsh, MS76, 5:25 Sat
Knap, Jaroslaw, CP1, 9:40 Wed
Krueger, Justin, MS169, 2:45 Mon
Knepley, Matthew G., MS181, 1:55 Mon
Ku, Seung-Hoe, MS201, 10:00 Tue
Knepley, Matthew G., PP204, 4:30 Mon
Kubatko, Ethan, MS206, 11:15 Tue
Knepley, Matthew, MS294, 2:00 Wed
Kuberry, Paul A., PP203, 4:30 Mon
Knight, Nicholas, MS38, 3:15 Sat
Kumar, Rakesh, PP2, 4:30 Sun
Koch, Zoe, MS172, 1:30 Mon
Kumar, Rakesh, CP23, 9:25 Wed
Koellermeier, Julian, PP105, 4:30 Sun
Kumari, Aradhana, PP8, 4:30 Sun
Koestler, Harald, MS124, 1:30 Sun
Kuo, Frances Y., MS159, 1:55 Mon
Koestler, Harald, MS239, 4:25 Tue
Kutz, J. Nathan, MS2, 10:15 Sat
Kohn, Robert V., MS40, 2:25 Sat
Kutz, J. Nathan, MS2, 10:15 Sat
Kolata, Bill, MS21, 2:25 Sat
Kutz, J. Nathan, MS28, 2:25 Sat
Kolata, Bill, MS47, 2:25 Sat
Kutz, J. Nathan, MS53, 4:35 Sat
Kolata, Bill, MS72, 2:25 Sat
Kuznetsov, Sergey V, CP4, 9:40 Wed
Kolata, William G., MS21, 10:15 Sat
Kyei, Yaw, CP23, 9:10 Wed
Kolata, William G., PD1, 12:15 Sat Kolata, William G., MS47, 2:25 Sat Kolata, William G., MS72, 4:35 Sat
L
Ladenheim, Scott, CP6, 9:10 Wed
Kolda, Tamara G., MS1, 10:15 Sat
Lafitte, Pauline, MS70, 4:35 Sat
Kolda, Tamara G., MS1, 10:40 Sat
Laguna, Ignacio, MS107, 2:20 Sun
Kolla, Hemanth, MS107, 2:45 Sun
Lahat, Dana, PP11, 4:30 Mon
Kong, Fande, CP6, 9:55 Wed
Laiu, Ming Tse P., PP105, 4:30 Sun
Kontos, Stavros, PP1, 4:30 Sun
Lambers, James V., MS5, 11:05 Sat
Lee, Junghoon, MS92, 10:25 Sun Lee, Lik Chuan, MS76, 5:00 Sat Lee, Young Ju, MS37, 3:40 Sat Lee, Young-Ju, MS11, 10:15 Sat Lee, Young-Ju, MS37, 2:25 Sat Lefantzi, Sophia, MS161, 1:55 Mon Leffell, Joshua I., PP6, 4:30 Sun Lehto, Erik, MS50, 3:15 Sat Leibs, Chris, MS137, 9:10 Mon Leiderman, Karin, MS22, 10:15 Sat Leiderman, Karin, MS122, 1:30 Sun Leiderman, Karin, MS140, 9:35 Mon Lermusiaux, Pierre F.J, MS218, 2:15 Tue Leung, Calvin, MS172, 2:02 Mon Leung, Vitus, MS262, 10:55 Wed Leung, Vitus, MS262, 10:55 Wed Leung, Vitus, MS287, 2:00 Wed LeVeque, Randall, PP102, 4:30 Sun Levy, Rachel, MS94, 9:10 Sun Levy, Rachel, MS119, 1:30 Sun Levy, Rachel, MS140, 10:00 Mon Levy, Rachel, MS222, 2:15 Tue Levy, Rachel, MS280, 2:25 Wed Lewicka, Marta, MS14, 11:30 Sat Lewis, Allison, MS218, 2:40 Tue Leyffer, Sven, PD0, 6:30 Sat
192
2015 SIAM Conference on Computational Science and Engineering
Leyffer, Sven, PD0, 8:30 Sat
Lignell, David O., MS20, 10:15 Sat
Ltaief, Hatem, MS118, 1:30 Sun
Leyffer, Sven, MS179, 1:30 Mon
Lignell, David O., MS20, 10:15 Sat
Ltaief, Hatem, MS145, 9:10 Mon
Leyffer, Sven, MS179, 1:30 Mon
Lim, Grace, MS172, 2:50 Mon
Ltaief, Hatem, MS171, 1:30 Mon
Leyffer, Sven, MS244, 4:25 Tue
Limogiannis, Nicolas, MS146, 10:30 Mon
Ltaief, Hatem, MS171, 1:55 Mon
Leyffer, Sven, MS269, 10:55 Wed
Lin, Fu, MS179, 2:45 Mon
Lu, Qiukai, MS121, 2:20 Sun
Li, Bing, MS7, 11:05 Sat
Lin, Fu, MS269, 10:55 Wed
Lu, Tiao, MS15, 11:05 Sat
Li, Fengyan, MS15, 10:15 Sat
Lin, Guang, MS213, 2:40 Tue
Lubin, Miles, MS246, 5:15 Tue
Li, Fengyan, MS15, 10:15 Sat
Lin, Junshan, MS44, 2:50 Sat
Luisier, Mathieu, MS209, 2:40 Tue
Li, Fengyan, MS64, 4:35 Sat
Lin, Lin, MS184, 10:00 Tue
Lukaczyk, Trent W., MS33, 3:40 Sat
Li, Fengyan, MS89, 9:10 Sun
Lin, Lin, MS209, 2:15 Tue
Luo, Songting, MS64, 4:35 Sat
Li, Fengyan, MS114, 1:30 Sun
Lin, Lin, MS234, 4:25 Tue
Li, Gaojin, MS96, 10:00 Sun
Lin, Ning, MS206, 10:25 Tue
Lupo Pasini, Massimiliano, MS210, 2:40 Tue
Li, Guanglian, PP8, 4:30 Sun
Lin, Paul, MS68, 5:00 Sat
Li, Harriet, MS91, 10:25 Sun
Linder, Christian, PP6, 4:30 Sun
Li, Hengguang, MS11, 11:30 Sat
Lindner, Florian, PP6, 4:30 Sun
Lushi, Enkeleida, MS122, 1:30 Sun
Li, Jichun, CP19, 10:10 Wed
Lipnikov, Konstantin, PP202, 4:30 Mon
Luszczek, Piotr, MS12, 10:15 Sat
Li, Jing-Rebecca, MS130, 10:25 Mon
Lischke, Anna, CP9, 9:10 Wed
Luszczek, Piotr, MS38, 2:25 Sat
Li, Longfei, MS32, 3:15 Sat
Litvinenko, Alexander, MS56, 4:35 Sat
Luszczek, Piotr, MS63, 4:35 Sat
Li, Matthew T., CP19, 9:25 Wed
Litvinenko, Alexander, PP5, 4:30 Sun
Luszczek, Piotr, MS235, 4:25 Tue
Li, Qin, MS120, 1:30 Sun
Litvinenko, Alexander, MS238, 5:15 Tue
Luttman, Aaron B., MS144, 9:10 Mon
Li, Ruo, MS248, 4:50 Tue
Liu, Hong, MS299, 2:00 Wed
Luttman, Aaron B., MS144, 9:10 Mon
Li, Shengtai, MS233, 5:15 Tue
Liu, Hong, MS299, 2:00 Wed
Luttman, Aaron B., MS170, 1:30 Mon
Li, Xiaofei, PP6, 4:30 Sun
Liu, James, MS265, 10:55 Wed
Lydon, Elizabeth, PP8, 4:30 Sun
Li, Xiaolin, MS208, 2:15 Tue
Liu, James, MS290, 2:00 Wed
Lynn, Brianna, PP7, 4:30 Sun
Li, Xiaolin, MS208, 2:15 Tue
Liu, Xing, MS63, 5:50 Sat
Lyon, Mark, MS66, 5:50 Sat
Li, Xiaolin, MS233, 4:25 Tue
Liu, Yanchao, MS269, 11:45 Wed
Lyon, Mark, MS154, 10:00 Mon
Li, Xiaolin, MS257, 10:55 Wed
Liu, Yuan, MS204, 10:00 Tue
Li, Xiaolin, MS282, 2:00 Wed
Liu, Yuan, MS204, 10:25 Tue
Li, Xiaoye Sherry, PP103, 4:30 Sun
Liu, Yuan, MS228, 2:15 Tue
Li, Xiaoye Sherry, MS149, 10:00 Mon
Liu, Yuan, MS251, 4:25 Tue
Ma, Yicong, MS37, 2:25 Sat
Li, Xiaoye Sherry, MS266, 10:55 Wed
Loeb, Andrew, PP15, 4:30 Mon
Mabuza, Sibusiso, PP1, 4:30 Sun
Li, Xiaoye Sherry, MS291, 2:00 Wed
Loffeld, John, MS186, 10:00 Tue
Maclachlan, Scott, MS261, 10:55 Wed
Li, Xingjie, MS23, 11:05 Sat
Loffeld, John, MS186, 10:00 Tue
Maclachlan, Scott, MS286, 2:00 Wed
Li, Xingjie, MS89, 10:25 Sun
Loffeld, John, MS210, 2:15 Tue
Madduri, Kamesh, MS250, 4:25 Tue
Li, Xingliang, MS268, 11:20 Wed
Logg, Anders, PP204, 4:30 Mon
Maeda, Hiroshi, MS38, 3:40 Sat
Li, Zhen, PP104, 4:30 Sun
Lolla, Tapovan, PP5, 4:30 Sun
Maggioni, Mauro, MS300, 2:50 Wed
Li, Zhen, MS242, 5:40 Tue
Long, Min, MS127, 2:20 Sun
Magruder, Caleb C., CP16, 10:10 Wed
Li, Zhijin, MS243, 4:25 Tue
Lopez, Anthony, MS198, 10:50 Tue
Mahadevan, Vijay, MS175, 1:30 Mon
Li, Zhilin, MS208, 3:30 Tue
Lopez, Florent, MS145, 9:35 Mon
Mahadevan, Vijay, PP204, 4:30 Mon
Liang, Chunlei, MS123, 1:55 Sun
Lotfian, Zahra S., PP14, 4:30 Mon
Mahoney, Michael, MS291, 2:00 Wed
Liao, Li, CP3, 9:25 Wed
Lotfian, Zahra S., CP18, 9:25 Wed
Main, Alex, MS57, 5:50 Sat
Lifflander, Jonathan, MS129, 10:00 Mon
Ltaief, Hatem, MS93, 9:10 Sun
Mainini, Laura, MS55, 5:00 Sat
Lushi, Enkeleida, MS96, 9:10 Sun Lushi, Enkeleida, MS96, 9:10 Sun
M
Ma, Yian, MS193, 10:25 Tue
2015 SIAM Conference on Computational Science and Engineering
193
Maischak, Matthias, MS278, 11:20 Wed
Mattis, Steven, MS84, 9:10 Sun
Messina, Paul C., PP14, 4:30 Mon
Malhotra, Dhairya, MS177, 2:45 Mon
Mattis, Steven, MS109, 1:30 Sun
Meyer, Francois G., MS29, 2:25 Sat
Mandli, Kyle T., MS102, 9:35 Sun
Mattis, Steven, MS109, 1:30 Sun
Meyer, Miriah, MS255, 11:20 Wed
Mandli, Kyle T., MS270, 10:55 Wed
Maxey, Martin, PP104, 4:30 Sun
Michalak, Anna, IP5, 8:15 Mon
Mandli, Kyle T., MS294, 2:00 Wed
May, Dave A., MS155, 10:00 Mon
Michalopoulou, Zoi-Heleni, PP7, 4:30 Sun
Mang, Andreas, MS214, 2:15 Tue
May, Elebeoba, MS13, 10:15 Sat
Miedlar, Agnieszka, CP25, 9:40 Wed
Mang, Andreas, MS214, 3:30 Tue
May, Elebeoba, MS39, 2:25 Sat
Mieussens, Luc, MS95, 10:25 Sun
Mang, Andreas, MS239, 4:25 Tue
May, Sandra, MS157, 1:55 Mon
Mignolet, Marc P., MS55, 5:50 Sat
Mannan, Forest O., PP1, 4:30 Sun
Mayo, Talea, MS166, 1:30 Mon
Mikler, Armin, MS100, 9:35 Sun
Manning, Cammey Cole, PD0, 6:30 Sat
Mayo, Talea, MS166, 2:45 Mon
Miller, Benjamin A., MS100, 9:10 Sun
Manning, Cammey Cole, PD0, 8:30 Sat
McClarren, Ryan G., MS19, 10:15 Sat
Miller, Benjamin A., MS100, 9:10 Sun
Mansour, Hassan, MS224, 2:40 Tue
McClarren, Ryan G., MS45, 2:25 Sat
Miller, Benjamin A., MS125, 1:30 Sun
Manteuffel, Thomas, MS85, 9:10 Sun
McClarren, Ryan G., MS70, 4:35 Sat
Miller, Eric, MS5, 10:15 Sat
Manteuffel, Thomas, MS110, 1:30 Sun
McClarren, Ryan G., MS95, 9:10 Sun
Miller, Eric, MS31, 2:25 Sat
Manteuffel, Thomas, MS110, 1:30 Sun
McClarren, Ryan G., MS120, 1:30 Sun
Millman, Kenneth J., MS78, 9:10 Sun
Manteuffel, Thomas, MS137, 9:10 Mon
McClarren, Ryan G., PP105, 4:30 Sun
Millman, Kenneth J., MS78, 9:10 Sun
Manzini, Gianmarco, PP202, 4:30 Mon
McClarren, Ryan G., MS147, 9:10 Mon
Min, MiSun, MS18, 10:15 Sat
Manzoni, Andrea, MS148, 9:10 Mon
McClarren, Ryan G., MS173, 1:30 Mon
Min, MiSun, MS44, 2:25 Sat
Manzoni, Andrea, MS148, 9:10 Mon
McClarren, Ryan, MS199, 10:00 Tue
Min, MiSun, MS89, 9:10 Sun
Manzoni, Andrea, MS174, 1:30 Mon
McClarren, Ryan, MS223, 2:15 Tue
Min, MiSun, MS114, 1:30 Sun
Mardal, Kent-Andre, MS229, 2:15 Tue
McCulloch, Andrew D., MS26, 10:15 Sat
Ming, Ju, MS282, 3:15 Wed
Mardal, Kent-Andre, MS252, 4:25 Tue
McDonald, Eleanor, MS229, 3:30 Tue
Minion, Michael, MS237, 4:50 Tue
Marquardt, Wolfgang, PD3, 12:15 Mon
McDonald, James, MS248, 5:15 Tue
Minkoff, Susan E., MS88, 9:10 Sun
Marques, Osni A., MS63, 4:35 Sat
Mcdougall, Damon, MS109, 2:20 Sun
Minkoff, Susan E., MS88, 9:10 Sun
Marques, Osni A., MS258, 10:55 Wed
McGraw, Carolyn, PP105, 4:30 Sun
Minkoff, Susan E., MS113, 1:30 Sun
Marques, Osni A., MS283, 2:00 Wed
Mcgregor, Duncan A., CP15, 9:10 Wed
Minkoff, Susan E., MS140, 9:10 Mon
Marshall, David, PD5, 12:45 Tue
McInnes, Lois Curfman, MS112, 1:30 Sun
Minkoff, Susan E., MS166, 1:30 Mon
Martin, Daniel, MS46, 2:25 Sat
McInnes, Lois Curfman, MS112, 1:30 Sun
Minkoff, Susan E., MS192, 10:00 Tue
Martin, Eileen R., PP14, 4:30 Mon
McInnes, Lois Curfman, MS131, 9:10 Mon
Minkoff, Susan E., MS216, 2:15 Tue
Martinsson, Gunnar, MS203, 10:25 Tue
McInnes, Lois Curfman, MS131, 10:00 Mon
Miras, Thomas, CP17, 9:25 Wed
Martinsson, Gunnar, MS291, 2:25 Wed
McInnes, Lois Curfman, MS247, 4:25 Tue
Misra, Satyajayant, MS192, 10:50 Tue
Mary, Theo, CP4, 9:55 Wed
McInnes, Lois Curfman, MS247, 4:25 Tue
Mitchell, Donna, MS216, 3:30 Tue
Marzouk, Youssef M., MS8, 10:15 Sat
Mead, Jodi, MS170, 2:45 Mon
Mitchell, John A., PP15, 4:30 Mon
Marzouk, Youssef M., MS34, 2:25 Sat
Medina, David, PP106, 4:30 Sun
Mitchell, Lawrence, MS207, 11:15 Tue
Marzouk, Youssef M., MS59, 4:35 Sat
Medina, Francis P., MS88, 10:25 Sun
Mitchell, William F., PP2, 4:30 Sun
Marzouk, Youssef M., MS132, 9:35 Mon
Medvinsky, Michael, MS36, 2:50 Sat
Mitchell, William F., CP26, 9:25 Wed
Marzouk, Youssef M., MS183, 10:00 Tue
Mehl, Miriam, PP14, 4:30 Mon
Mitrano, Arthur, MS126, 2:45 Sun
Masse, Danielle D., PP4, 4:30 Sun
Meidani, Hadi, CP5, 9:55 Wed
Moe, Scott, PP10, 4:30 Mon
Mathelin, Lionel, MS55, 5:25 Sat
Meister, Oliver, MS270, 10:55 Wed
Mohamed, Mamdouh S., PP9, 4:30 Mon
Matthews, Christopher, MS259, 11:45 Wed
Meixner, Jessica, MS230, 3:30 Tue
Mohammad, Zakerzadeh, MS150, 9:35 Mon
Matthies, Hermann, MS56, 4:35 Sat
Meng, Xiong, MS23, 11:30 Sat
Mohseni, Kamran, MS30, 3:40 Sat
Matthysen, Roel, PP11, 4:30 Mon
Menhorn, Friedrich, PP5, 4:30 Sun
Molavi Tabrizi, Amirhossein, PP7, 4:30 Sun
194
2015 SIAM Conference on Computational Science and Engineering
Molzahn, Daniel, MS269, 11:20 Wed
Najm, Habib N., MS161, 1:30 Mon
Nguyen, Thien Binh, MS293, 2:25 Wed
Monk, Peter B., MS278, 12:10 Wed
Nakajima, Kengo, MS17, 10:15 Sat
Nicholls, David P., MS18, 10:15 Sat
Morales Escalante, Jose A., CP12, 9:40 Wed
Nakajima, Kengo, MS17, 10:40 Sat
Nicholls, David P., MS44, 2:25 Sat
Morel, Jim E., MS156, 1:55 Mon
Nakajima, Kengo, MS43, 2:25 Sat
Nicholls, David P., MS44, 3:40 Sat
Morii, Youhi, PP14, 4:30 Mon
Nakajima, Kengo, MS68, 4:35 Sat
Nielsen, Eric, MS86, 10:25 Sun
Morikuni, Keiichi, MS275, 11:20 Wed
Nance, James, PP6, 4:30 Sun
Nobile, Fabio, MS106, 1:30 Sun
Morlighem, Mathieu, MS71, 5:00 Sat
Narayan, Akil, MS191, 11:15 Tue
Nordaas, Magne, MS252, 5:15 Tue
Morrisey, Thomas, PP15, 4:30 Mon
Narayanan, Sri Hari Krishn, MS190, 10:00 Tue
Norton, Richard A., PP5, 4:30 Sun
Narayanan, Sri Hari Krishn, MS190, 10:00 Tue
Nouy, Anthony, MS264, 11:45 Wed
Morzfeld, Matthias, MS8, 10:15 Sat Moser, Dieter, MS286, 2:00 Wed
Narcowich, Francis J., MS50, 2:50 Sat
Moser, Robert D., MS165, 1:30 Mon
Nashed, Youssef, PP14, 4:30 Mon
O
Motamed, Mohammed, MS277, 11:45 Wed
Naumov, Maxim, MS163, 2:45 Mon
Moulton, David, MS112, 2:20 Sun
Nave, Jean-Christophe, MS66, 4:35 Sat
Moxey, David, MS67, 5:50 Sat
Navon, Ionel M., MS168, 1:30 Mon
Moxey, David, MS103, 9:10 Sun
Navon, Ionel M., MS168, 1:30 Mon
Moxey, David, MS128, 1:30 Sun
Navon, Ionel M., MS194, 10:00 Tue
Moxey, David, MS231, 3:05 Tue
Navon, Ionel M., MS218, 2:15 Tue
Mu, Lin, MS265, 11:20 Wed
Navon, Ionel M., MS243, 4:25 Tue
Mubayi, Anuj, MS192, 10:00 Tue
Navon, Ionel M., MS267, 10:55 Wed
Olivares, Nicole, MS119, 2:34 Sun
Mubayi, Anuj, MS192, 11:15 Tue
Navon, Ionel M., MS292, 2:00 Wed
Oliver, Todd, MS139, 9:10 Mon
Mueller, Juliane, MS60, 4:35 Sat
Neckel, Tobias, MS34, 2:50 Sat
Mueller, Juliane, MS60, 4:35 Sat
Neckel, Tobias, MS80, 9:10 Sun
Oliver, Todd, MS161, 2:45 Mon
Müller, Benjamin, MS85, 9:10 Sun
Neckel, Tobias, MS105, 1:30 Sun
Olson, Luke, PP205, 4:30 Mon
Mundani, Ralf-Peter, MS294, 3:15 Wed
Nedich, Angelia, MS104, 1:30 Sun
Munoz, Francisco, MS269, 10:55 Wed
Olson, Luke, MS254, 4:25 Tue
Negri, Federico, MS174, 1:30 Mon
Olson, Luke, MS254, 4:25 Tue
Munson, Todd, MS269, 10:55 Wed
Neilan, Michael J., PP207, 4:30 Mon
Olson, Sarah D., MS99, 10:00 Sun
Münzenmaier, Steffen, MS110, 2:45 Sun
Nemati Hayati, Arash, PP206, 4:30 Mon
Olvera De La Cruz, Monica, MS40, 2:50 Sat
Murillo, Michael, MS199, 10:50 Tue
Nemec, Marian, MS205, 10:50 Tue
Ommen, Jürgen, PP11, 4:30 Mon
Müthing, Steffen, PP204, 4:30 Mon
Nestler, Franziska, CP20, 9:40 Wed
O’Neill, Ben, MS286, 3:15 Wed
Mycek, Paul, MS138, 9:10 Mon
Neupane, Prapti, CP23, 9:55 Wed
Myers, Aaron, PP13, 4:30 Mon
Newman, Chris, MS147, 10:00 Mon
Myers, Andrew, MS186, 10:00 Tue
Nguyen, Cuong, MS62, 4:35 Sat
Myers, Andrew, MS186, 11:15 Tue
Nguyen, Cuong, MS87, 9:10 Sun
Myers, Andrew, MS210, 2:15 Tue
Nguyen, Cuong, MS87, 9:10 Sun
Morrison, Rebecca, MS74, 5:25 Sat Morrison, Rebecca, MS166, 2:20 Mon
Nguyen, Hoa, MS13, 10:15 Sat
Nourgaliev, Robert, MS220, 2:40 Tue
O’Connell, Meghan, MS143, 10:25 Mon Oden, J. Tinsley, PD3, 12:15 Mon Oesper, Layla, MS125, 1:55 Sun Ohi, Yoshiharu, PP10, 4:30 Mon Ohshima, Satoshi, MS43, 3:15 Sat Okamoto, Naoya, PP9, 4:30 Mon O’Leary, Patrick, MT2, 4:30 Mon O’Leary, Patrick, PD1, 12:15 Sat
Oliver, Todd, MS165, 1:30 Mon
O’Neill, Kristin, MS21, 10:15 Sat O’Neill, Kristin, MS47, 2:25 Sat O’Neill, Kristin, MS72, 4:35 Sat Onwunta, Akwum, MS252, 5:40 Tue Opsomer, Peter, PP10, 4:30 Mon Orban, Dominique, MS9, 10:15 Sat
N
Nguyen, Hoa, MS16, 10:15 Sat
Nadal-Quiros, Monica, PP12, 4:30 Mon
Nguyen, Hoa, MS16, 10:15 Sat
Nadarajah, Siva, MS98, 10:00 Sun
Nguyen, Hoa, MS39, 2:25 Sat
Ortan, Alexandra, PP14, 4:30 Mon
Nagy, James G., MS170, 2:20 Mon
Nguyen, Hoa, MS42, 2:25 Sat
Osborn, Sarah, MS138, 10:25 Mon
Najm, Habib N., MS135, 9:10 Mon
Nguyen, Hoang-Ngan, MS22, 10:15 Sat
Oseledets, Ivan, MS224, 3:05 Tue
Najm, Habib N., MS135, 9:10 Mon
Nguyen, Hoang-Ngan, MS22, 10:15 Sat
Osher, Stanley J., MS2, 11:05 Sat
Orban, Dominique, MS9, 11:05 Sat Orban, Dominique, MS35, 2:25 Sat
2015 SIAM Conference on Computational Science and Engineering
195
Osting, Braxton, MS3, 10:15 Sat
Perego, Mauro, MS71, 5:25 Sat
Platte, Rodrigo B., MS180, 1:55 Mon
Osting, Braxton, MS29, 2:25 Sat
Perego, Mauro, MS193, 10:00 Tue
Plechac, Petr, MS164, 1:30 Mon
Otten, Matthew, MS18, 11:05 Sat
Perego, Mauro, MS217, 2:15 Tue
Plews, Julia A., PP14, 4:30 Mon
Ouaknin, Gaddiel, CP1, 9:55 Wed
Perego, Mauro, MS242, 4:25 Tue
Polizzi, Eric, MS184, 10:00 Tue
Owen, J. Michael, MS167, 4:25 Tue
Perez, Fernando, MS182, 11:15 Tue
Polizzi, Eric, MS209, 2:15 Tue
Owhadi, Houman, IP7, 8:15 Tue
Perez-Arancibia, Carlos A., MS18, 11:30 Sat
Polizzi, Eric, MS234, 4:25 Tue
Oxberry, Geoffrey M., MS285, 2:25 Wed
Perline, Kyle, PP13, 4:30 Mon
Polizzi, Eric, MS273, 10:55 Wed
Ozyilmaz, Emre, PP15, 4:30 Mon
Persson, Per-Olof, MS123, 2:45 Sun
Polizzi, Eric, MS297, 2:00 Wed
Pestana, Jennifer, MS83, 9:10 Sun
Pollock, Sara, MS11, 11:05 Sat
P
Pestana, Jennifer, MS108, 1:30 Sun
Porcelli, Margherita, MS229, 2:40 Tue
Pain, Christopher, MS267, 11:45 Wed
Pestana, Jennifer, MS108, 1:30 Sun
Pothen, Alex, MS256, 10:55 Wed
Palacios, Francisco, MS302, 2:00 Wed
Peterka, Tom, MS255, 11:45 Wed
Pothen, Alex, MS281, 2:00 Wed
Pan, Wenxiao, PP104, 4:30 Sun
Peters, Bas, MS35, 2:25 Sat
Pothen, Alex, MS281, 2:25 Wed
Pan, Wenxiao, MS193, 10:00 Tue
Peters, Michael, MS185, 11:15 Tue
Poulson, Jack L., MS12, 11:30 Sat
Pan, Wenxiao, MS193, 10:00 Tue
Peterson, Euguenia, PP15, 4:30 Mon
Poulson, Jack L., PP204, 4:30 Mon
Pan, Wenxiao, MS217, 2:15 Tue
Peterson, Jacob, MS223, 3:30 Tue
Powell, Catherine, PP13, 4:30 Mon
Pan, Wenxiao, MS242, 4:25 Tue
Peterson, Kara, PP203, 4:30 Mon
Proctor, Joshua, MS2, 10:15 Sat
Panagiotou, Eleni, MS114, 1:30 Sun
Peterson, Kara, MS296, 3:15 Wed
Proctor, Joshua, MS28, 2:25 Sat
Panda, Nishant, MS109, 1:55 Sun
Petiton, Serge G., MS17, 10:15 Sat
Proctor, Joshua L., MS28, 2:25 Sat
Panda, Nishant, PP101, 4:30 Sun
Petiton, Serge G., MS43, 2:25 Sat
Proctor, Joshua, MS53, 4:35 Sat
Panoff, Robert M., MS232, 4:25 Tue
Petiton, Serge G., MS68, 4:35 Sat
Proft, Jennifer, MS206, 10:00 Tue
Panoff, Robert M., MS232, 4:25 Tue
Petiton, Serge G., MS68, 4:35 Sat
Proft, Jennifer, MS230, 2:15 Tue
Papadimitriou, Costas, MS158, 2:45 Mon
Petra, Cosmin G., MS195, 11:15 Tue
Proft, Jennifer, MS230, 2:15 Tue
Park, Hyeongkae, MS173, 1:30 Mon
Petra, Noemi, MS46, 2:25 Sat
Prokopenko, Andrey, CP8, 9:10 Wed
Parno, Matthew, PP13, 4:30 Mon
Petra, Noemi, MS71, 4:35 Sat
Pulch, Roland, MS183, 11:15 Tue
Parno, Matthew, MS215, 2:40 Tue
Petra, Noemi, MS71, 5:50 Sat
Pask, John, MS184, 10:50 Tue
Petzold, Linda R., SP1, 8:30 Sat
Pasquale, Laura, PP14, 4:30 Mon
Petzold, Linda R., MS115, 2:45 Sun
Pathmanathan, Pras, MS51, 3:40 Sat
Pfluger, Dirk, MS185, 10:50 Tue
Patra, Abani K., MS139, 9:10 Mon
Phillips, Cynthia, PD1, 12:15 Sat
Patra, Abani K., MS165, 1:30 Mon
Phillips, Edward, MS295, 2:50 Wed
Paul-Dubois-Taine, Arthur, MS30, 2:50 Sat
Phillips, Jeff, PD4, 12:45 Tue
Pawlak, Wojciech, PP1, 4:30 Sun
Phipps, Eric, PP106, 4:30 Sun
Pawlowski, Roger, MS142, 10:00 Mon
Phipps, Eric, MS138, 9:10 Mon
Payne, Joshua, MS304, 2:25 Wed
Phipps, Eric, MS138, 9:35 Mon
Peherstorfer, Benjamin, MS4, 10:15 Sat
Phipps, Eric, MS164, 1:30 Mon
Peherstorfer, Benjamin, MS30, 2:25 Sat
Pilosov, Michael, PP14, 4:30 Mon
Qiu, Jingmei, MS147, 9:10 Mon
Peherstorfer, Benjamin, MS55, 4:35 Sat
Pinar, Ali, MS195, 10:00 Tue
Qiu, Jingmei, MS173, 1:30 Mon
Peherstorfer, Benjamin, MS260, 12:10 Wed
Pinar, Ali, MS219, 2:15 Tue
Qiu, Jingmei, MS199, 10:00 Tue
Pellegrini, Francois, MS287, 2:00 Wed
Pippig, Michael, PP6, 4:30 Sun
Qiu, Jingmei, MS223, 2:15 Tue
Peña, Antonio J., MS235, 5:15 Tue
Piret, Cecile M., MS24, 11:30 Sat
Qiu, Weifeng, MS62, 5:25 Sat
Peng, Jun, MS251, 4:50 Tue
Piret, Cecile M., MS154, 10:25 Mon
Qu, Zhilin, MS51, 2:25 Sat
peng, Zhangli, MS242, 4:50 Tue
Plank, Gernot, MS26, 11:30 Sat
Quaife, Bryan D., MS286, 2:25 Wed
Q Qiu, Jingmei, MS19, 10:15 Sat Qiu, Jingmei, MS45, 2:25 Sat Qiu, Jingmei, MS70, 4:35 Sat Qiu, Jingmei, MS89, 9:10 Sun Qiu, Jingmei, MS95, 9:10 Sun Qiu, Jingmei, MS120, 1:30 Sun Qiu, Jingmei, MS114, 1:30 Sun Qiu, Jingmei, PP105, 4:30 Sun
Quenneville-Belair, Vincent, PP2, 4:30 Sun
196
2015 SIAM Conference on Computational Science and Engineering
R
Ricketson, Lee F., MS223, 3:05 Tue
Ruthotto, Lars, MS116, 1:30 Sun
Rabidoux, Scott, PP14, 4:30 Mon
Ridzal, Denis, PP203, 4:30 Mon
Ruthotto, Lars, MS143, 9:10 Mon
Rachh, Manas, MS177, 1:55 Mon
Ridzal, Denis, MS274, 10:55 Wed
Ruthotto, Lars, MS169, 1:30 Mon
Radunskaya, Ami, MS113, 1:30 Sun
Rieger, Christian, MS126, 1:55 Sun
Ruthotto, Lars, MS214, 3:05 Tue
Radunskaya, Ami, MS113, 2:45 Sun
Ritter, Otto, MS117, 2:20 Sun
Ruthotto, Lars, MS246, 4:25 Tue
Ragan-Kelley, Min, PP204, 4:30 Mon
Rizzi, Francesco, MS160, 1:55 Mon
Ruthotto, Lars, MS246, 4:50 Tue
Raghavan, Padma, PD3, 12:15 Mon
Roberts, Nathan, PP204, 4:30 Mon
Ruuth, Steven, MS237, 5:15 Tue
Ragusa, Jean C., PP2, 4:30 Sun
Robinson, Allen C., MS156, 2:20 Mon
Ryan, Sarah M., MS219, 2:15 Tue
Ragusa, Jean C., MS284, 2:25 Wed
Rodrigo, Carmen, MS202, 11:15 Tue
Ryckelynck, David, MS30, 3:15 Sat
Rai, Prashant, MS213, 3:05 Tue
Roe, Philip L., MS176, 1:30 Mon
Rycroft, Chris H., MS220, 3:30 Tue
Rajamanickam, Siva, MS141, 9:35 Mon
Rogers, Ryan M., PP11, 4:30 Mon
Rajamanickam, Siva, MS200, 10:00 Tue Rajamanickam, Siva, MS250, 4:25 Tue
Rognes, Marie E., MS207, 10:50 Tue Rognes, Marie E., MS229, 2:15 Tue
S
Saad, Tony, PP206, 4:30 Mon
Ramachandran, Prabhu, MS126, 2:20 Sun
Rognes, Marie E., MS252, 4:25 Tue
Raman, Barani, MS54, 5:25 Sat
Rohrle, Oliver, MS85, 10:25 Sun
Ramet, Pierre, MS145, 9:10 Mon
Roman, Bogdan, MS276, 12:10 Wed
Rana, Anirudh Singh, MS248, 5:40 Tue
Romero, Daniel, MS192, 10:00 Tue
Ranjan, Harsh, PP9, 4:30 Mon
Rossmanith, James A., MS271, 10:55 Wed
Rappoport, Juri M., PP14, 4:30 Mon
Rossmanith, James A., MS271, 10:55 Wed
Rastigejev, Yevgenii, CP24, 9:25 Wed
Rossmanith, James A., MS295, 2:00 Wed
Ratnaswamy, Vishagan, PP13, 4:30 Mon
Rostami, Minghao W., PP9, 4:30 Mon
Rave, Stephan, MS236, 4:25 Tue
Rouet, Francois-Henry, MS291, 2:50 Wed
Ravindran, S.S., MS148, 10:00 Mon
Rowley, Clarence, MS260, 11:20 Wed
Ray, Jaideep, MS161, 2:20 Mon
Rozza, Gianluigi, MS168, 2:45 Mon
Ray, Navamita, MS97, 9:35 Sun
Rozza, Gianluigi, MS187, 10:00 Tue
Reeves, Daniel B., PP12, 4:30 Mon
Rozza, Gianluigi, MS187, 10:00 Tue
Reid, Andrew, PP14, 4:30 Mon
Rozza, Gianluigi, MS211, 2:15 Tue
Reinarz, Anne, MS119, 1:46 Sun
Rozza, Gianluigi, MS236, 4:25 Tue
Relton, Samuel, PP11, 4:30 Mon
Rozza, Gianluigi, MS260, 10:55 Wed
Renaut, Rosemary A., MS31, 3:40 Sat
Safta, Cosmin, MS195, 10:00 Tue
Rozza, Gianluigi, MS285, 2:00 Wed
Rennich, Steven C., MS163, 1:30 Mon
Safta, Cosmin, MS219, 2:15 Tue
Rudi, Johann, PP9, 4:30 Mon
Rennich, Steven C., MS163, 1:30 Mon
Ruede, Ulrich J., MS80, 9:10 Sun
Resch, Michael, MS118, 1:55 Sun
Ruede, Ulrich J., MS105, 1:30 Sun
Sahni, Onkar, MS225, 3:30 Tue
Ruede, Ulrich J., MS131, 9:10 Mon
Saibaba, Arvind, MS5, 10:15 Sat
Ruede, Ulrich J., MS131, 9:10 Mon
Saibaba, Arvind, MS5, 11:30 Sat
Ruede, Ulrich J., MS197, 10:00 Tue
Saibaba, Arvind, MS31, 2:25 Sat
Ruede, Ulrich J., MS221, 2:15 Tue
Sakkaplangkul, Puttha, PP10, 4:30 Mon
Rukavishnikov, Viktor, CP19, 9:10 Wed
Salac, David, MS139, 9:10 Mon
Runborg, Olof, MS301, 3:15 Wed
Salac, David, MS139, 10:00 Mon
Reshniak, Viktor, CP22, 9:55 Wed Restrepo, Juan M., MS218, 3:05 Tue Rey, Thomas, MS173, 2:20 Mon Reynolds, Daniel R., MS175, 2:20 Mon Rhebergen, Sander, MS155, 9:10 Mon Rhebergen, Sander, MS155, 9:10 Mon Rhebergen, Sander, MS181, 1:30 Mon Richfield, Owen, PP9, 4:30 Mon Ricketson, Lee F., PP10, 4:30 Mon
Rupard, Morgan, PP14, 4:30 Mon Rupp, Karl, PP204, 4:30 Mon Ruthotto, Lars, MS91, 9:10 Sun
Saad, Tony, MS304, 2:50 Wed Saad, Yousef, MS68, 5:25 Sat Saad, Yousef, MS273, 10:55 Wed Saad, Yousef, MS297, 2:00 Wed Sachs, Ekkehard W., MS91, 10:00 Sun Sachse, Frank B., MS26, 10:15 Sat Sachse, Frank B., MS26, 10:40 Sat Sachse, Frank B., MS51, 2:25 Sat Sadanand, Chandrika, PP8, 4:30 Sun Sadayappan, P, MS25, 11:30 Sat Sadeghitohidi, Ana, PP11, 4:30 Mon Sadek, Carol, MS146, 9:10 Mon Sadre-Marandi, Farrah, PP10, 4:30 Mon Safro, Ilya, MS226, 3:30 Tue Safta, Cosmin, MS195, 10:00 Tue
Sahni, Onkar, MS201, 10:00 Tue Sahni, Onkar, MS225, 2:15 Tue
Salac, David, MS165, 1:30 Mon Salgado, Abner J., MS48, 3:15 Sat
2015 SIAM Conference on Computational Science and Engineering
197
Salgado, Abner J., MS178, 1:55 Mon
Sayas, Francisco J. J., MS278, 10:55 Wed
Shankar, Varun, MS75, 4:35 Sat
Salinger, Andrew, MS225, 2:40 Tue
Sayas, Francisco J. J., MS303, 2:00 Wed
Shankar, Varun, MS101, 9:10 Sun
Salles, Nicolas, MS278, 10:55 Wed
Saye, Robert, MS130, 9:35 Mon
Shankar, Varun, MS126, 1:30 Sun
Salles, Nicolas, MS278, 10:55 Wed
Schärer, Roman P., PP105, 4:30 Sun
Shao, Meiyue, MS61, 4:35 Sat
Salles, Nicolas, MS303, 2:00 Wed
Schatz, Martin D., MS25, 10:15 Sat
Shao, Meiyue, MS61, 5:50 Sat
Salomon, Julien, MS236, 4:50 Tue
Schatz, Martin D., MS25, 10:15 Sat
Shashkov, Mikhail, MS156, 1:30 Mon
Saltzman, Jeffrey, PD1, 12:15 Sat
Schiavazzi, Daniele E., MS189, 11:15 Tue
Shashkov, Mikhail, MS156, 1:30 Mon
Saltzman, Jeffrey, MS92, 9:10 Sun
Schilders, Wil, PD0, 8:30 Sat
Shashkov, Mikhail, MS167, 4:25 Tue
Saltzman, Jeffrey, MS92, 9:10 Sun
Schlottbom, Matthias, MS19, 11:05 Sat
Shen, Han-Wei, MS255, 10:55 Wed
Saltzman, Jeffrey, MS117, 1:30 Sun
Schmidt, John A., MS129, 9:10 Mon
Shen, Han-Wei, MS255, 12:10 Wed
Samatova, Nagiza, IP1, 9:00 Sat
Schmidt, Kathleen, MS135, 10:25 Mon
Shen, Jiguang, MS62, 5:00 Sat
Samulyak, Roman, CP23, 9:40 Wed
Schneider, Kai, MS41, 2:50 Sat
Sanchez-Uribe, Manuel A., PP203, 4:30 Mon
Schneider, Reinhold, MS56, 5:25 Sat
Sanchez-Uribe, Manuel A., MS282, 2:50 Wed
Schroder, Jacob B., MS261, 10:55 Wed
Sanchez-Vizuet, Tonatiuh, MS303, 2:00 Wed
Schroder, Jacob B., MS286, 2:00 Wed
Sandu, Adrian, MS136, 9:10 Mon Sandu, Adrian, MS136, 9:10 Mon Sandu, Adrian, MS168, 1:30 Mon Sandu, Adrian, MS162, 1:30 Mon Sandu, Adrian, MS194, 10:00 Tue Sandu, Adrian, MS188, 10:00 Tue Sandu, Adrian, MS218, 2:15 Tue
Schofield, Elizabeth, MS222, 2:40 Tue Schroder, Jacob B., MS261, 11:20 Wed Schroeder, Chris, MS201, 10:50 Tue Schroeder, Will, IP6, 11:20 Mon Schulthess, Thomas C., MS12, 10:40 Sat Schulthess, Thomas C., MS118, 1:30 Sun Schulz, Martin, MS82, 10:25 Sun Schulz, Volker H., MS69, 5:00 Sat Schwab, Christoph, MS81, 9:35 Sun Schwab, Christoph, MS215, 2:15 Tue
Shephard, Mark S., MS97, 9:10 Sun Shephard, Mark S., MS121, 1:30 Sun Sherwin, Spencer, MS103, 9:10 Sun Sherwin, Spencer, MS128, 1:30 Sun Sherwin, Spencer, MS150, 9:10 Mon Shiflet, Angela B., MS232, 4:25 Tue Shiflet, George W., MS232, 4:50 Tue Shinar, Tamar, MS122, 2:20 Sun Shirokoff, David, MS41, 2:25 Sat Shirokoff, David, MS41, 3:15 Sat Shirokoff, David, MS66, 4:35 Sat Shoemaker, Christine A., MS60, 4:35 Sat Shoemaker, Christine A., MS60, 5:50 Sat Shu, Chi-Wang, MS157, 2:20 Mon
Sandu, Adrian, MS212, 2:15 Tue
Schwartz, Fernando, PP4, 4:30 Sun
Sandu, Adrian, MS243, 4:25 Tue
Schwartz, Fernando, CP26, 9:55 Wed
Sandu, Adrian, MS237, 4:25 Tue
Schwendeman, Donald W., MS32, 3:40 Sat
Sandu, Adrian, MS267, 10:55 Wed
Scovazzi, Guglielmo, MS57, 5:25 Sat
Sandu, Adrian, MS292, 2:00 Wed
Seal, David C., MS120, 2:20 Sun
Santosa, Fadil, MS44, 2:25 Sat
Sego, Landon H., MS54, 4:35 Sat
Saraswat, Jyoti, CP14, 9:10 Wed
Sego, Landon H., MS54, 4:35 Sat
Sargsyan, Khachik, MS135, 9:10 Mon
Seibold, Benjamin, MS70, 5:00 Sat
Sargsyan, Khachik, MS161, 1:30 Mon
Seidel, Ed, PD3, 12:15 Mon
Sargsyan, Khachik, MS189, 10:00 Tue
Semiyari, Hamid, PP10, 4:30 Mon
Slaughter, Andrew, MS259, 10:55 Wed
Sargsyan, Khachik, MS213, 2:15 Tue
Sen, Arindam, PP9, 4:30 Mon
Slaughter, Andrew, MS284, 2:00 Wed
Sargsyan, Khachik, MS238, 4:25 Tue
Senter, Michael, MS172, 2:34 Mon
Slepcev, Dejan, MS29, 2:50 Sat
Sargsyan, Khachik, MS264, 11:20 Wed
Seol, E. Seegyoung, MS97, 10:25 Sun
Sloan, Ian H., MS159, 1:30 Mon
Sariaydin, Selin, MS169, 1:55 Mon
Shankar, Sadasivan, MS65, 5:00 Sat
Slota, George, MS250, 5:40 Tue
Saunders, Michael A., MS35, 3:15 Sat
Shankar, Varun, MS24, 10:15 Sat
Sayadi, Taraneh, MS86, 9:10 Sun
Shankar, Varun, MS24, 10:40 Sat
Smith, Cameron, MS97, 9:10 Sun
Sayas, Francisco J. J., MS208, 3:05 Tue
Shankar, Varun, MS50, 2:25 Sat
Smith, Cameron, MS121, 1:30 Sun
Shumlak, Uri, MS271, 12:10 Wed Siebenborn, Martin, MS229, 3:05 Tue Siefert, Christopher, PP203, 4:30 Mon Silva, Claudio T., MS255, 10:55 Wed Sime, Nathan, CP3, 9:10 Wed Simons, Julie, MS113, 2:20 Sun Simpson, Olivia, PP3, 4:30 Sun Slattery, Stuart, MS210, 2:15 Tue
Smetana, Kathrin, CP18, 10:10 Wed
198
2015 SIAM Conference on Computational Science and Engineering
Smith, Cameron, PP103, 4:30 Sun
Stefanescu, Razvan, MS168, 1:30 Mon
Sutton, Oliver, PP202, 4:30 Mon
Smith, Cameron, MS200, 10:00 Tue
Stefanescu, Razvan, MS168, 1:55 Mon
Svyatskiy, Daniil, PP2, 4:30 Sun
Smith, Ralph C., MS33, 3:15 Sat
Stefanescu, Razvan, MS194, 10:00 Tue
Swiler, Laura, MS164, 2:20 Mon
Smith, Ralph C., PD5, 12:45 Tue
Stefanescu, Razvan, MS218, 2:15 Tue
Sylvand, Guillaume, MS171, 2:45 Mon
Soane, Ana Maria, CP14, 9:25 Wed
Stefanescu, Razvan, MS243, 4:25 Tue
Szyld, Daniel B., CP25, 9:10 Wed
Socha, Katherine, MS280, 2:50 Wed
Stefanescu, Razvan, MS267, 10:55 Wed
Sockwell, Chad, PP15, 4:30 Mon
Stefanescu, Razvan, MS292, 2:00 Wed
Solonen, Antti, MS33, 2:50 Sat
Steffen, Kyle R., MS10, 11:05 Sat
Taddei, Tommaso, MS285, 2:50 Wed
Somayajula, Sangeetha, MS117, 1:55 Sun
Stein, David, MS122, 1:55 Sun
Taitano, William T., PP105, 4:30 Sun
Steiner, Johannes, MS79, 9:10 Sun
Taitano, William T., MS199, 11:15 Tue
Steiner, Johannes, MS79, 9:10 Sun
Takahashi, Daisuke, MS283, 2:50 Wed
Steinmetz, Philipp, PP15, 4:30 Mon
Takaki, Nick, PP14, 4:30 Mon
Sticko, Simon, MS36, 3:15 Sat
Takaki, Tomohiro, MS124, 2:45 Sun
Still, Charles (Bert) H., MS279, 10:55 Wed
Takhtaganov, Timur, MS298, 2:50 Wed
Still, Charles (Bert) H., MS304, 2:00 Wed
Takizawa, Hiroyuki, MS258, 12:10 Wed
Stinchcombe, Adam, MS96, 10:25 Sun
Talnikar, Chaitanya, MS111, 2:45 Sun
Stinis, Panos, PP104, 4:30 Sun
Tamellini, Lorenzo, MS215, 3:05 Tue
Stobb, Michael T., CP20, 9:55 Wed
Tang, Hansong, CP16, 9:55 Wed
Stogner, Roy, MS165, 1:55 Mon
Tang, Qi, MS204, 10:00 Tue
Spiteri, Raymond J., MS162, 1:30 Mon
Stoyanov, Miroslav, MS134, 10:00 Mon
Tang, Qi, MS228, 2:15 Tue
Spiteri, Raymond J., MS188, 10:00 Tue
Strogies, Nikolai, PP15, 4:30 Mon
Tang, Qi, MS251, 4:25 Tue
Spiteri, Raymond J., MS212, 2:15 Tue
Strogies, Nikolai, CP14, 9:55 Wed
Tang, Qi, MS271, 10:55 Wed
Spiteri, Raymond J., MS237, 4:25 Tue
Stronz, James A., MS146, 9:42 Mon
Tang, Qi, MS271, 11:45 Wed
Spiteri, Raymond J., MS237, 4:25 Tue
Strzodka, Robert, MS221, 3:30 Tue
Tang, Qi, MS295, 2:00 Wed
Springer, Daryl J., MS50, 3:40 Sat
Sudret, Bruno, MS189, 10:00 Tue
Tanguy, Sebastien, MS245, 4:50 Tue
Srinivasan, S, PP11, 4:30 Mon
Sun, Andy, MS219, 2:40 Tue
Tani, Mattia, MS83, 10:25 Sun
Stadler, Georg, MS8, 10:15 Sat
Sun, Pengtao, MS37, 2:50 Sat
Tartakovsky, Alexander, MS193, 11:15 Tue
Stadler, Georg, MS46, 2:25 Sat
Sundahl, Bryan E., PP14, 4:30 Mon
Tausch, Johannes, MS278, 11:45 Wed
Stadler, Georg, MS34, 2:25 Sat
Sundar, Hari, MS62, 4:35 Sat
Teckentrup, Aretha, MS133, 9:35 Mon
Stadler, Georg, MS71, 4:35 Sat
Sundar, Hari, MS62, 4:35 Sat
Temam, Roger M., MS268, 10:55 Wed
Stadler, Georg, MS59, 4:35 Sat
Sundar, Hari, MS87, 9:10 Sun
Tempone, Raul F., MS191, 10:50 Tue
Stadler, Georg, MS59, 4:35 Sat
Sundnes, Joakim, MS76, 4:35 Sat
Tendulkar, Saurabh, MS121, 1:30 Sun
Stafford, Shane, MS284, 2:00 Wed
Sundnes, Joakim, MS76, 4:35 Sat
ter Maten, E. Jan W., MS162, 2:20 Mon
Starinshak, David, MS167, 4:50 Tue
Surowiec, Thomas M., MS298, 2:00 Wed
Teran, Joseph, MS196, 10:50 Tue
Stark, Philip B., MS78, 9:10 Sun
Sussman, Daniel L., MS125, 2:20 Sun
Teranishi, Keita, MS82, 9:10 Sun
Starke, Gerhard, MS85, 9:10 Sun
Sussman, Mark, MS196, 10:00 Tue
Teranishi, Keita, MS107, 1:30 Sun
Starke, Gerhard, MS85, 9:35 Sun
Sussman, Mark, MS220, 2:15 Tue
Teranishi, Keita, MS134, 9:10 Mon
Starke, Gerhard, MS110, 1:30 Sun
Sussman, Mark, MS245, 4:25 Tue
Teranishi, Keita, MS160, 1:30 Mon
Starke, Gerhard, MS137, 9:10 Mon
Sussman, Mark, MS245, 4:25 Tue
Terashima, Hiroshi, PP9, 4:30 Mon
Stathopoulos, Andreas, MS297, 2:50 Wed
Sutherland, James C., MS20, 11:05 Sat
Terejanu, Gabriel, MS139, 9:35 Mon
Steck, Sebastian, MS174, 2:20 Mon
Sutherland, James C., MS129, 10:25 Mon
Terrasse, Isabelle, MS151, 10:25 Mon
Sorensen, Danny C., MS28, 3:15 Sat Sosonkina, Masha, MS61, 5:00 Sat Sousedik, Bedrich, MS264, 12:10 Wed Spantini, Alessio, MS8, 11:30 Sat Spector, Michael, MS299, 2:00 Wed Spector, Michael, MS299, 2:25 Wed Spencer, Benjamin, MS259, 12:10 Wed Spicer, Amy, MS94, 9:42 Sun Spiller, Elaine, MS165, 2:45 Mon Spiteri, Raymond J., MS136, 9:10 Mon
T
2015 SIAM Conference on Computational Science and Engineering
199
Terrel, Andy R., MT2, 10:25 Mon
Turcksin, Bruno, MS207, 10:25 Tue
Van Straalen, Brian, PP103, 4:30 Sun
Terzi, Evamaria, MS1, 11:30 Sat
Turkiyyah, George M., PP102, 4:30 Sun
Van Straalen, Brian, MS200, 10:00 Tue
Tezaur, Radek, MS48, 2:50 Sat
Turner, Peter R., MS146, 9:10 Mon
Van ‘t Wout, Elwin, MS303, 3:15 Wed
Thompson, Terese, PP9, 4:30 Mon
Turner, Peter R., MS172, 1:30 Mon
Vandereycken, Bart, MS224, 2:15 Tue
Thornock, Jeremy, PP206, 4:30 Mon
Turner, Peter R., MS280, 2:00 Wed
Vandereycken, Bart, MS224, 2:15 Tue
Thornton, Anthony R., MS181, 1:30 Mon
Turner, Peter R., MS280, 2:00 Wed
Vandereycken, Bart, MS249, 4:25 Tue
Thrush, Brandon D., MS146, 9:58 Mon Thurston, Courtney E., PP7, 4:30 Sun Tian, Xiaochuan, PP10, 4:30 Mon Tobin, William R., MS250, 4:25 Tue Tokman, Mayya, MS136, 10:25 Mon Tomov, Stanimire, MS12, 10:15 Sat Tomov, Stanimire, MS38, 2:25 Sat Tomov, Stanimire, MS63, 4:35 Sat Tomov, Vladimir, MS279, 11:20 Wed Tornberg, Anna-Karin, MS22, 10:40 Sat Toroczkai, Zoltan, MS1, 11:05 Sat Torrilhon, Manuel, MS19, 11:30 Sat Toth, Alexander R., PP14, 4:30 Mon Toth, Gyula I., CP7, 9:10 Wed Townsend, Alex, MS180, 2:20 Mon Tranquilli, Paul, MS136, 10:00 Mon Trask, Nathaniel, MS193, 10:00 Tue Trask, Nathaniel, MS217, 2:15 Tue Trask, Nathaniel, MS217, 2:15 Tue Trask, Nathaniel, MS242, 4:25 Tue Traub, Thomas, MS303, 2:25 Wed Treister, Eran, CP2, 9:55 Wed Trout, Charlotte M., MS232, 5:40 Tue Tryggvason, Gretar, MS52, 4:35 Sat Tryggvason, Gretar, MS52, 4:35 Sat Tsang, Alan Cheng Hou, MS96, 9:35 Sun Tsilifis, Panagiotis, PP208, 4:30 Mon
Varduhn, Vasco, CP11, 9:55 Wed
U
Udagedara, Indika G., MS144, 10:25 Mon Uddameri, Elma A., MS216, 2:15 Tue Uddameri, Elma A., MS216, 3:05 Tue Udell, Madeleine R., PP11, 4:30 Mon Uekermann, Benjamin, MS6, 11:05 Sat Ufimtsev, Vladimir, MS125, 1:30 Sun Ulbrich, Michael, MS252, 4:25 Tue
Vaughan, Courtenay T., CP11, 9:10 Wed Vazquez-Gonzalez, Thibaud, PP1, 4:30 Sun Vecharynski, Eugene, MS273, 11:45 Wed Veerapaneni, Shravan, MS151, 10:00 Mon Venkatasubramanian, Vaithianathan, MS256, 10:55 Wed Venkatasubramanian, Vaithianathan, MS256, 10:55 Wed
Ullmann, Sebastian, PP13, 4:30 Mon
Venkatasubramanian, Vaithianathan, MS281, 2:00 Wed
Ullrich, Paul, MS272, 10:55 Wed
Vermeire, Brian C., MS98, 9:10 Sun
Ullrich, Paul, MS272, 10:55 Wed
Vermeire, Brian C., MS123, 1:30 Sun
Ullrich, Paul, MS296, 2:00 Wed
Vermeire, Brian C., MS123, 2:20 Sun
Uminsky, David T., MS140, 9:10 Mon
Vermeire, Brian C., MS150, 9:10 Mon
Uminsky, David T., MS140, 10:25 Mon
Vermeire, Brian C., MS176, 1:30 Mon
Urban, Karsten, MS236, 5:15 Tue
Veroy-Grepl, Karen, MS211, 3:30 Tue
Urombo, Jack, PP10, 4:30 Mon
Vervliet, Nico, MS249, 5:40 Tue
Uryasev, Stan, MS298, 2:25 Wed
Vidal-Codina, Ferran, MS183, 10:50 Tue
V
Villa, Umberto E., MS164, 1:55 Mon Vincent, Peter E., MS98, 9:10 Sun
Vaaland, Uno B., CP13, 9:40 Wed
Vincent, Peter E., MS98, 9:10 Sun
Valeev, Edward F., MS25, 11:05 Sat
Vincent, Peter E., MS123, 1:30 Sun
Valiathan, Chandni, MS92, 9:35 Sun
Vincent, Peter E., MS150, 9:10 Mon
Van Bloemen Waanders, Bart G., PP14, 4:30 Mon
Vincent, Peter E., MS176, 1:30 Mon
Van Bloemen Waanders, Bart G., MS244, 4:50 Tue
Vogelius, Michael S., MS65, 5:25 Sat
Vishnampet, Ramanathan, PP9, 4:30 Mon
Van Bloemen Waanders, Bart G., MS274, 10:55 Wed
Vogl, Chris, MS257, 11:45 Wed Vokt, Joseph, PP11, 4:30 Mon
Tu, Xuemin, MS290, 2:50 Wed
Van Bloemen Waanders, Bart G., MS298, 2:00 Wed
Tuma, Miroslav, MS275, 11:45 Wed
van Gennip, Yves, MS29, 3:40 Sat
Vollmer, Charlie, PP5, 4:30 Sun
Tuncer, Ozan, MS262, 11:45 Wed
Van Huffel, Sabine, IP8, 11:50 Tue
Voth, Gregory, PP104, 4:30 Sun
Turc, Catalin, MS18, 10:40 Sat
van Leeuwen, Tristan, MS35, 3:40 Sat
Vuduc, Richard, MS27, 3:15 Sat
Turcksin, Bruno, PP106, 4:30 Sun
Van Loan, Charles, MS25, 10:40 Sat
Vuik, Kees, CP6, 9:25 Wed
Tu, Jonathan H., MS99, 9:10 Sun Tu, Jonathan H., MS99, 9:35 Sun Tu, Shuang Z., MS176, 2:20 Mon
Van Straalen, Brian, MS127, 1:55 Sun
Vogman, Genia, CP12, 9:10 Wed Volfovsky, Alexander, PP13, 4:30 Mon
200
2015 SIAM Conference on Computational Science and Engineering
W
Watson, Jean-Paul, MS195, 10:00 Tue
Wilson, Greg, MT3, 2:15 Tue
Wachtel, Andreas, MS119, 2:50 Sun
Watson, Jean-Paul, MS219, 2:15 Tue
Wilson, Greg, MT4, 4:25 Tue
Walker, Homer F., MS142, 10:25 Mon
Webster, Clayton G., MS81, 9:10 Sun
Wirkus, Stephen, MS88, 9:10 Sun
Wall, Samuel, MS76, 4:35 Sat
Webster, Clayton G., MS106, 1:30 Sun
Wirkus, Stephen, MS113, 1:30 Sun
Wall, Samuel, MS76, 5:50 Sat
Webster, Clayton G., MS106, 2:20 Sun
Wirkus, Stephen, MS140, 9:10 Mon
Walsh, Scott, MS84, 10:00 Sun
Webster, Clayton G., MS133, 9:10 Mon
Wirkus, Stephen, MS166, 1:30 Mon
Walther, Andrea, MS69, 4:35 Sat
Webster, Clayton G., MS159, 1:30 Mon
Wirkus, Stephen, MS192, 10:00 Tue
Walther, Andrea, MS69, 5:50 Sat
Webster, Clayton G., PP201, 4:30 Mon
Wirkus, Stephen, MS216, 2:15 Tue
Waluga, Christian, MS93, 9:35 Sun
Webster, Clayton G., MS185, 10:00 Tue
Witherden, Freddie, MS128, 2:45 Sun
Wang, Jialei, PP11, 4:30 Mon
Wei, Ermin, MS104, 2:20 Sun
Wittmann, Roland, MS273, 11:20 Wed
Wang, Jianxun, CP13, 9:10 Wed
Wei, Qi-Huo, MS42, 3:15 Sat
Wojciechowski, Keith, PP2, 4:30 Sun
Wang, Junping, PP207, 4:30 Mon
Weile, Daniel, MS303, 2:50 Wed
Wolf, Michael, MS200, 11:15 Tue
Wang, Junping, MS290, 2:25 Wed
Weinzierl, Tobias, MS270, 10:55 Wed
Wolfson-Pou, Jordi, MS197, 10:25 Tue
Wang, Kainan, MS8, 11:05 Sat
Weinzierl, Tobias, MS270, 12:10 Wed
Womeldorff, Geoff, CP11, 9:40 Wed
Wang, Li, MS70, 5:25 Sat
Weinzierl, Tobias, MS294, 2:00 Wed
Wong, Christopher, MS94, 9:58 Sun
Wang, Lu, MS226, 3:05 Tue
Weiser, Martin, MS79, 10:00 Sun
Wong, Elizabeth, MS179, 1:55 Mon
Wang, Mengdi, MS104, 2:45 Sun
Weiss, Robert, MS270, 11:45 Wed
Wong, Kwai L., CP3, 9:40 Wed
Wang, Pochuan, CP7, 9:25 Wed
Weller, Hilary, MS130, 10:00 Mon
Wong, Tony E., PP13, 4:30 Mon
Wang, Qing, PP12, 4:30 Mon
Wen, Ci, PP15, 4:30 Mon
Wong, Yau Shu, MS268, 12:10 Wed
Wang, Qiqi, MS86, 9:10 Sun
Weng, Tsui-Wei, PP5, 4:30 Sun
Woodring, Jonathan, MT1, 12:10 Wed
Wang, Qiqi, MS111, 1:30 Sun
Westerink, Joannes, MS206, 10:50 Tue
Woodring, Jonathan, MT1, 9:10 Sun
Wang, Qiqi, MS202, 10:50 Tue
Whitaker, Ross, MS54, 5:00 Sat
Woodring, Jonathan, MT2, 1:30 Sun
Wang, Ting, PP13, 4:30 Mon
White, Ryan, MS94, 10:30 Sun
Woodward, Carol S., PD0, 6:30 Sat
Wang, Weichung, MS283, 3:15 Wed
Whitehead, Jared P., MS293, 2:00 Wed
Woodward, Carol S., PP103, 4:30 Sun
Wang, Yanqiu, MS290, 3:15 Wed
Wild, Stefan, MS60, 5:25 Sat
Woodward, Carol S., MS142, 9:35 Mon
Wang, Yaohong, CP21, 9:10 Wed
Wildey, Tim, PP13, 4:30 Mon
Woodward, Carol S., MS162, 2:45 Mon
Wang, Ying, MS114, 1:55 Sun
Willcox, Karen E., MS53, 5:50 Sat
Woodward, Paul R., MS279, 10:55 Wed
Wang, Yuhang, PP12, 4:30 Mon
Willcox, Karen E., MS131, 9:10 Mon
Woopen, Michael, MS87, 10:25 Sun
Wang, Zhijun, PP12, 4:30 Mon
Willcox, Karen E., MS131, 9:35 Mon
Wortmann, Daniel, MS184, 11:15 Tue
Warburton, Tim, MS43, 2:25 Sat
Willert, Jeffrey A., MS19, 10:15 Sat
Wright, Grady B., MS24, 10:15 Sat
Warburton, Timothy, MS103, 10:25 Sun
Willert, Jeffrey A., MS45, 2:25 Sat
Wright, Grady B., MS50, 2:25 Sat
Willert, Jeffrey A., MS70, 4:35 Sat
Wright, Grady B., MS75, 4:35 Sat
Willert, Jeffrey A., MS95, 9:10 Sun
Wright, Grady B., MS101, 9:10 Sun
Willert, Jeffrey A., MS120, 1:30 Sun
Wright, Grady B., MS126, 1:30 Sun
Willert, Jeffrey A., PP105, 4:30 Sun
Wrobel, Jacek, MS282, 2:00 Wed
Wathen, Andrew J., MS155, 9:10 Mon
Willert, Jeffrey A., MS147, 9:10 Mon
Wu, Lingfei, PP15, 4:30 Mon
Wathen, Andrew J., MS181, 1:30 Mon
Willert, Jeffrey A., MS173, 1:30 Mon
Wu, Lingfei, CP25, 9:25 Wed
Wathen, Andy, MS83, 9:10 Sun
Willert, Jeffrey A., MS199, 10:00 Tue
Wu, Yuqi, MS55, 4:35 Sat
Wathen, Andy, MS83, 9:10 Sun
Willert, Jeffrey A., MS223, 2:15 Tue
Wathen, Andy, MS108, 1:30 Sun
Wilson, Anastasia B., PP12, 4:30 Mon
Watkins, Daniel, PP13, 4:30 Mon
Wilson, Greg, MT3, 4:30 Mon
Watson, Cody, MS172, 1:46 Mon
Wilson, Greg, MT4, 4:30 Mon
Ward, Joseph, MS50, 2:25 Sat Ward, Rachel, MS106, 2:45 Sun Ward, Rachel, MS213, 3:30 Tue Wasilkowski, Grzegorz W., MS81, 10:25 Sun
X Xiao, Feng, MS196, 10:00 Tue Xiao, Feng, MS220, 2:15 Tue
2015 SIAM Conference on Computational Science and Engineering
201
Xiao, Feng, MS220, 2:15 Tue
Yano, Masayuki, MS187, 10:25 Tue
Zhang, Linan, MS172, 2:18 Mon
Xiao, Feng, MS245, 4:25 Tue
Ye, Xiu, MS265, 10:55 Wed
Zhang, Mengping, MS228, 2:15 Tue
Xie, Xiaoping, MS265, 12:10 Wed
Yetkin, Emrullah Fatih, MS134, 9:35 Mon
Zhang, Qinghai, MS245, 5:40 Tue
Xin, Zixing, MS241, 5:15 Tue
Yin, Wotao, MS104, 1:30 Sun
Zhang, Shangyou, MS265, 11:45 Wed
Xing, Yulong, PP105, 4:30 Sun
Yin, Wotao, MS104, 1:55 Sun
Zhang, Shun, MS152, 9:10 Mon
Xing, Yulong, MS257, 10:55 Wed
Ying, Lexing, MS27, 3:40 Sat
Zhang, Shun, MS178, 1:30 Mon
Xiu, Dongbin, MS191, 10:00 Tue
Yokota, Rio, MS177, 1:30 Mon
Zhang, Shun, MS178, 2:45 Mon
Xiu, Dongbin, MS191, 10:00 Tue
Yonkee, Nathan, CP3, 9:55 Wed
Zhang, Weijian, MS246, 5:40 Tue
Xiu, Dongbin, MS215, 2:15 Tue
Yoo, Yeonjoo, PP12, 4:30 Mon
Zhang, Xiangxiong, MS49, 3:15 Sat
Xiu, Dongbin, MS240, 4:25 Tue
Yordanov, Pencho, PP208, 4:30 Mon
Zhang, Xu, MS152, 9:35 Mon
Xiu, Dongbin, MS264, 10:55 Wed
Yordanov, Pencho, CP5, 9:25 Wed
Zhang, Xu, MS282, 2:25 Wed
Yoshimura, Shinobu, IP3, 8:15 Sun
Zhang, Yanzhi, CP20, 9:25 Wed
Younes, Laurent, MS214, 2:15 Tue
Zhang, Yin, MS249, 5:15 Tue
Yu, Haijun, MS64, 5:25 Sat
Zhang, Yongtao, MS204, 10:00 Tue
Yu, Hui, MS15, 11:30 Sat
Zhang, Zheng, PP5, 4:30 Sun
Yu, Lijun, PP12, 4:30 Mon
Zhang, Zheng, MS289, 3:15 Wed
Yu, Yue, MS57, 5:00 Sat
Zhao, Lin, PP9, 4:30 Mon
Yu, Yue, MS114, 2:45 Sun
Zheng, Mengdi, MS240, 4:50 Tue
Yue, Pengtao, CP21, 9:25 Wed
zheng, Wen, MS220, 3:05 Tue
Xiu, Dongbin, MS289, 2:00 Wed Xu, Jinchao, PP104, 4:30 Sun Xu, Kun, MS64, 5:00 Sat Xu, Ling, MS114, 2:20 Sun Xu, Xiaowen, CP8, 9:40 Wed Xu, Yangyang, MS249, 4:50 Tue Xu, Yuanwei, PP13, 4:30 Mon Xu, Zhengfu, MS251, 4:25 Tue Xu, Zhiliang, MS204, 10:50 Tue Xu, Zhiliang, MS233, 4:50 Tue
Y
Yakovlev, Sergey B., MS87, 9:35 Sun Yamanaka, Akinori, MS124, 2:20 Sun Yamazaki, Ichitaro, MS291, 3:15 Wed Yan, Bokai, MS120, 2:45 Sun Yan, Jue, MS233, 5:40 Tue Yang, Chao, MS184, 10:00 Tue Yang, Chao, MS209, 2:15 Tue Yang, Chao, MS234, 4:25 Tue Yang, He, MS49, 2:25 Sat Yang, Ulrike M., PP103, 4:30 Sun Yang, Ulrike M., MS149, 9:10 Mon Yang, Ulrike M., MS175, 1:30 Mon Yang, Ulrike Meier, MS149, 9:10 Mon Yang, Xiu, MS189, 10:25 Tue Yang, Yang, MS23, 10:15 Sat Yang, Yang, MS49, 2:25 Sat Yang, Yang, MS49, 2:50 Sat Yang, Yong, PP2, 4:30 Sun Yang, Zhang, CP7, 9:55 Wed
Zhong, Xinghui, MS49, 3:40 Sat
Z
Zhou, Aihui, MS234, 4:25 Tue
Zabaras, Nicholas, MS191, 10:25 Tue
Zhou, Beckett, MS302, 2:50 Wed
Zabaras, Nicholas, CP5, 9:10 Wed
Zhou, Dong, CP18, 9:40 Wed
Zabaras, Nicholas, CP13, 9:25 Wed
Zhou, Tao, MS189, 10:50 Tue
Zaharatos, Brian, MS58, 5:50 Sat
Zhou, Tao, MS240, 5:15 Tue
Zahr, Matthew J., MS4, 11:05 Sat
Zhou, Yongcheng, MS257, 11:20 Wed
Zakerzadeh, Seyed Hamed, CP10, 9:10 Wed
Zhu, Anna, MS94, 10:14 Sun
Zander, Elmar, MS238, 5:40 Tue
Zhu, Hongyu, MS292, 2:25 Wed
Zavala, Victor, MS244, 4:25 Tue
Zhu, Xueyu, MS289, 2:25 Wed
Zeng, Xianyi, CP18, 9:55 Wed
Zhu, Yunrong, MS11, 10:15 Sat
Zepeda-Núñez, Leonardo, MS119, 1:30 Sun
Zhu, Yunrong, MS11, 10:15 Sat
Zhang, Fan, PP14, 4:30 Mon
Zhu, Yunrong, MS37, 2:25 Sat
Zhang, Guannan, MS81, 9:10 Sun
Zimmermann, Ralf, MS30, 2:25 Sat
Zhang, Guannan, MS106, 1:30 Sun
Zorin, Denis, MS227, 2:40 Tue
Zhang, Guannan, MS133, 9:10 Mon
Zosso, Dominique, MS3, 10:15 Sat
Zhang, Guannan, MS133, 10:25 Mon
Zosso, Dominique, MS29, 2:25 Sat
Zhang, Guannan, MS159, 1:30 Mon
Zulehner, Walter, MS229, 2:15 Tue
Zhang, Guannan, PP201, 4:30 Mon
Zupanski, Milija, MS243, 4:50 Tue
Zhang, Guannan, MS185, 10:00 Tue
Zwicknagl, Barbara, MS101, 9:10 Sun
Zhang, Hong, MS141, 10:25 Mon Zhang, Hong, MS212, 3:05 Tue Zhang, Hongxuan, MS254, 5:40 Tue
202
2015 SIAM Conference on Computational Science and Engineering
Notes
2015 SIAM Conference on Computational Science and Engineering
203
Conference Budget SIAM Conference on Computational Science & Engineering March 14 - 18, 2015 Salt Lake City, UT Expected Paid Attendance: 1374 Revenue Registration Income
Total
$463,255 $463,255
Expenses Printing Organizing Committee Invited Speakers Food and Beverage AV Equipment/ Room Rental and Telecommunication Advertising Conference Labor (including benefits) Professional Services (Recording / Mobile App / Child Care) Other (supplies, staff travel, freight, misc.) Administrative Accounting/Distribution & Shipping Information Systems Customer Service Marketing Office Space (Building) Other SIAM Services Total:
$12,200 $6,500 $42,600 $148,700 $92,000 $12,000 $101,710 $27,700 $11,000 $46,042 $24,551 $44,267 $16,720 $ , $26,262 $16,611 $17,544 $646,407
Net Conference Expense:
($183,152)
Support Provided by SIAM:
$183,152 $0
Estimated Support for Travel Awards not included above: Early Career and Students 83
$54,750
The Calvin L. Rampton Salt Palace Convention Center Floor Plan