[IEEE 2012 IEEE 20th International Symposium on Modelling, Analysis [PDF]

Feb 19, 2017 - ... Technology Rochester, New York 14623 Email: [email protected] Shuaiwen Song SCAPE laboratory Virginia T

1 downloads 6 Views 317KB Size

Recommend Stories


IEEE International Symposium on Biomedical Imaging
You're not going to master the rest of your life in one day. Just relax. Master the day. Than just keep

2018 IEEE MTT-S International Microwave Symposium
At the end of your life, you will never regret not having passed one more test, not winning one more

2017 IEEE International Symposium on Antennas and Propagation and
You can never cross the ocean unless you have the courage to lose sight of the shore. Andrè Gide

2016 IEEE International Symposium on Phased Array Systems and Technology
You have to expect things of yourself before you can do them. Michael Jordan

IEEE Std 1310-2012
You often feel tired, not because you've done too much, but because you've done too little of what sparks

4TH IEEE Smart Cities Symposium
Nothing in nature is unbeautiful. Alfred, Lord Tennyson

2018 IEEE International Geoscience and Remote Sensing Symposium 2018 IEEE International
You're not going to master the rest of your life in one day. Just relax. Master the day. Than just keep

Neural Networks, 1988., IEEE International Conference on
No amount of guilt can solve the past, and no amount of anxiety can change the future. Anonymous

IEEE Transactions on Computers
Before you speak, let your words pass through three gates: Is it true? Is it necessary? Is it kind?

Education, IEEE Transactions on
Open your mouth only if what you are going to say is more beautiful than the silience. BUDDHA

Idea Transcript


(https://documents.tips/register.html)

(https://documents.tips/)

HOME (HTTPS://DOCUMENTS.TIPS/) LEADERSHIP (HTTPS://DOCUMENTS.TIPS/CATEGORY/LEADERSHIP-MANAGEMENT.HTML) TECHNOLOGY (HTTPS://DOCUMENTS.TIPS/CATEGORY/TECHNOLOGY.HTML) EDUCATION (HTTPS://DOCUMENTS.TIPS/CATEGORY/EDUCATION.HTML) MORE TOPICS (HTTPS://DOCUMENTS.TIPS/CATEGORY.HTML)

/

Home (https://documents.tips/) / Documents (https://documents.tips/category/documents.html) [IEEE 2012 IEEE 20th International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - Washington, DC, USA (2012.08.7-2012.08.9)] 2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems - Energy-Aware Replica Selection for Data-Intensive Services in Cloud (https://documents.tips/documents/ieee-2012-ieee-20th-international-symposium-on-modelling-analysissimulation-58aa27d7cb226.html)

[IEEE 2012 IEEE 20th International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - Washington, DC, USA (2012.08.7-2012.08.9)] 2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems - EnergyAware Replica Selection for Data-Intensive Services in Cloud REPORT (HTTPS://DOCUMENTS.TIPS/REPORT-COPYRIGHT/IEEE-2012-IEEE-20TH-INTERNATIONALDocuments (https://documents.tips/category/documents.html)

SYMPOSIUM-ON-MODELLING-ANALYSIS-SIMULATION-58AA27D7CB226/5750A3401A28ABCF0CA14DA7) Please download to view (https://documents.tips/download/link/ieee-2012-ieee-20th-international-symposium-on-modellinganalysis-simulation-58aa27d7cb226) Energy-Aware Replica Selection for Data-Intensive Services in Cloud Bo Li Department of Computer Science Rochester Institute of Technology Rochester, New York 14623 Email: [email protected] Shuaiwen Song SCAPE laboratory Virginia

RECOMMENDED (https://documents.tips/documents/ieee-2012ieee-20th-international-symposium-on-modellinganalysis-simulation-589dde30a8445.html)

replica selection system is designed and developed to collect energy consumption information of data centers. The results

[IEEE 2012 IEEE 20th International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - Washington, DC, USA (2012.08.7-2012.08.9)] 2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems - Solving the TCP-Incast Problem with Application-Level Scheduling (https://documents.tips/documents/ieee2012-ieee-20th-internationalsymposium-on-modelling-analysissimulation-589dde30a8445.html)

show that the total energy cost can be effectively reduced by using our decentralized replica selection system comparing

Documents

Tech Blacksburg, VA 24060 Email: [email protected] Ivona Bezakova Department of Computer Science Rochester Institute of Technology Rochester, New York 14623 Email: [email protected] Kirk W. Cameron SCAPE laboratory Virginia Tech Blacksburg, VA 24060 Email: [email protected] Abstract—With the increasing energy cost in data centers, an energy efficient approach to provide data intensive services in the cloud is highly in demand. This paper solves the energy cost reduction problem of data centers by formulating an energy- aware replica selection problem in order to guide the distribution of workload among data centers. The current popular centralized replica selection approaches address such problem but they lack scalability and are vulnerable to a crash of the central coordinator. Also, they do not take total data center energy cost as the primary optimization target. We propose a simple decentralized replica selection system implemented with two distributed optimization algorithms (consensus-based distributed projected subgradient method and Lagrangian dual decomposi- tion method) to work with clients as a decentralized coordinator. We also compare our energy-aware replica selection approach with the replica selection where a round-robin algorithm is implemented. A prototype of the decentralized

with a round-robin method. It also has low calculation and communication overhead and can be easily adapted to the real world cloud environment. I. INTRODUCTION In the cloud, some services are replicated in geographically different data centers. If the clients want to use such services, they need to specify one or more data centers to connect with. The problem of how to choose from those data centers is called replica selection. The way in which replicas are selected for

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2012ieee-20th-international-symposium-on-modellinganalysis-simulation-589eb9576a6e9.html)

failure problem. Decentralized coordinator architecture performs much better in terms of scalability and reliability [3].

[IEEE 2012 IEEE 20th International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - Washington, DC, USA (2012.08.7-2012.08.9)] 2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems - Comparing the ns-3 Propagation Models (https://documents.tips/documents/ieee2012-ieee-20th-internationalsymposium-on-modelling-analysissimulation-589eb9576a6e9.html)

However, previous work on designing decentralized replica selection systems have not considered total system energy cost

Documents

as the primary optimization target. In this work, we propose an energy-aware decentralized replica selection system for

(https://documents.tips/category/documents.html)

client requests is important because choosing the right replicas clients can help users achieve the optimal performance, such as minimal latency, the least packet loss, or maximal available bandwidth. Also, the service providers can benefit from more balanced load among different replicas to minimize the operating cost. Since energy consumption has comprised a significant part of the costs in data centers [1], energy cost reduction becomes an essential way to reduce and minimize the operating cost of data centers. The electricity price is one of the factors affecting this cost in data centers. It varies with different locations and times in a day. Therefore, an energy-aware replica selection system considering real time electricity price is highly demanded by cloud service providers in order to reduce the total energy cost. A replica selection system with distributed architecture is better than the centralized setup in terms of reliability and scalability. Centralized architecture, such as MapReduce[2], implements a centralized coordinator to handle and distribute all the tasks. However, its scalability issue causes bottlenecks in handling a large amount of client requests. Also it is not reliable because of the single point

data-intensive applications in the cloud. It can reduce the total energy cost by distributing the data-intensive workload among all the data centers. An energy consumption model for data centers is built to indicate the relationship between workload and energy consumption for data-intensive applications. This model does not involve the cooling energy

(https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation-585f9e27655c4.html)

cloud, we build a correlation between energy consumption and the requests from clients. We consider the power

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems - Effect of Latent Errors on the Reliability of Data Storage Systems (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation-585f9e27655c4.html)

consumption in the data center coming from two major parts: the server nodes, and the network infrastructure. Since data-

Documents

consumption because cooling system varies among different data centers. We adapt the Lagrangian dual decomposition method (LDDM) to our replica selection sys- tem to solve this global optimization problem in parallel. We compare both performance and total energy cost of our approach with those of consensus-based distributed projected subgradient method (CDPSM)[4] and baseline round-robin replica selection method. II. METHODOLOGY Some important notations in this section are summarized in Table I. They are mapped to the system architecture in Fig. 1. TABLE I NOTATIONS C Set of all clients N Set of replicas pcn Traffic load mapped from client c to replica n Pn Constraint sets on replica n Bn Bandwidth capacity on replica n Rc Traffic load of the request from client c un Unit price (¢) of power in replica n an Weight value of replica n in consensus-based algorithm n, n Weight scalars for the energy consumption of computer devices and network devices in replica n 2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems 1526-7539/12 $26.00 © 2012 IEEE DOI 10.1109/MASCOTS.2012.66 504 A. Energy Consumption Model for Data Center In order to minimize the energy cost of data-intensive applications in the

intensive applications in the cloud are mainly disk intensive (e.g. video streaming and sharing), we make the assumption that there is a linear relationship between the workload and the energy consumption of a server in the cloud [5]. The relationship between the energy consumption of the network infrastructure(such as routers and switches) and workload is determined by the technologies used in designing the hardware. For the equipments which use Dynamic Voltage Scaling

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation-589d2e76d544b.html)

architecture is shown in Fig. 1. In the system, each replica keeps listening to the clients’ requests. Once the requests to

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems Networks of Order Independent Queues with Signals (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation-589d2e76d544b.html)

the replica selection system are received, these replicas will start to cooperate with each other to solve the global

Documents

(DVS) [6] as the energy reduction approach at archi- tecture level, the relationship between traffic load and energy consumption can be modeled as cubic. For instance, Ethernet interface cards applying DVS and DFS (Dynamic Frequency Scaling), have been proved to support this relationship [7]. Therefore, we can get a weighted combination of linear (for servers) and cubic (for network devices) relationship between energy consumption and network traffic load in our model. The total energy consumption of all the replicas can be modeled as: Eg = ∑ n un · (n ∑ c pcn + n( ∑ c pcn) 3) (1) where and are weight scalars. The goal of our problem is to minimize Eg for the clients’ requests to the data centers. The global optimization problem can be formulated as: minimize pcn Eg = ∑ n En subject to fn(P ) = ∑ c pcn −Bn ≤ 0, "n Î N hc(P ) = ∑ n pcn −Rc = 0, "c Î C (2) where En = un · (n ∑ c pcn + n( ∑ c pcn) 3) is the energy consumption in replica n, fn(P ) is the bandwidth capacity constraint of replica n, hc(P ) is the request constraint of client c. The problem turns out to be a cubic objective function with several linear equality and inequality constraints. B. Decentralized Replica Selection System The decentralized replica selection system is built on the infrastructure of data centers without any additional devices. The

optimization problem. Fig. 1. Distributed services in replicas We investigate two distributed algorithms to solve the global optimization problem in parallel. 1) Consensus-based Distributed Projected Subgradient Method (CDPSM): This method is originally proposed to solve constrained optimization problems in multi-agents net- works [8]. In our paper, we adapt this method to our de- centralized replica selection system. The objective function Eg in our replica selection problem is the

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation-589f01a8deb06.html)

formulate the Lagrangian dual problem from the global optimization problem as: 505

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems - Impact of Multi-access Links on the Internet Topology Modeling (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation-589f01a8deb06.html)

minimize pcn L(pcn, µ) = N∑ n=1 En + C∑ c=1 µi · hc(P ) subject to fn(P ) = ∑ c pcn −Bn ≤ 0, "n Î N (4) Fig. 2 shows the

Documents

sum of functions which are local objective functions for replicas in the form of Eg = ∑ n En. Each replica works on solving its own local optimization problem En which is subject to the local constraints pcn Î Pn, where Pn is a subset of the constraint sets that have local variables of replica n . The optimization problem in replica n can be formulated as: minimize pcn En subject to pncn Î Pn The consensus mechanism can combine solution of sub- problems to form the global optimization solution. Given pcn is the solution to the global optimization problem, each replica n starts by estimating {pcn | c Î C n Î N}n Î Pn and updating its solution pcn iteratively by cooperating with other replicas. The consensus and projection procedure for itera- tively estimating can be denoted by the following equation: pncn(k + 1) = ProjPn [ N∑ j=1 ajn · pjcn(k)− dk · gn(k)]+ (3) where ajn are the weights of all the replicas, dk > 0 is the step size, and gn(k) is the subgradient on its local objective function En. ProjPn [p * cn] + = arg min pcnÎPn p *cn − pcn 2) Lagrangian Dual Decomposition Method (LDDM): Since there are dependencies in the global variables among replicas, we need to decouple them in order to solve the prob- lem in parallel. LDDM provides us with a way to solve such problem. Given the original problem (2), we can

comparison of simulated convergence rates of these two methods. We do the simulation work with three replicas using MatLab. For solving the same optimization problem, the CDPSM converges slower than the LDDM. So theoretically, for our energy-aware replica selection problem, the LDDM is expected to have higher performance. Fig. 2. Simulation results III. EXPERIMENT RESULTS The model parameters used in the experiments in this section are defined in Table II. We also

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation.html)

data-intensive applications. In the best case scenario, the total energy cost using LDDM can be reduced by 17.8%

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems Performance Analysis of SoftwareDefined Networking (SDN) (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation.html)

comparing with a round-robin method and 15.3% comparing with CDPSM. In future, we plan to further improve LDDM

Documents

set the value of scalars an,n, and n in Table I to be 1. In our experiments, we use TABLE II PARAMETERS SETUP IN THE MODEL Replica 1 2 3 4 5 6 7 8 Band Cap (MB/s) 100 100 100 100 100 100 100 100 two types of data-intensive applications: the video streaming and the distributed file service, running on a power-aware HPC cluster ’SystemG’ [9] at Virginia Tech. The size per request is different for these two applications. We set the size per request for the video streaming is approximately 100 MBytes and for the distributed file service it is approximately 10 MBytes. The power consumption cost results are shown in Fig. 3 4 5 6. Fig. 3. Energy cost of each replica for video streaming application Fig. 4. Total energy cost of all the replicas for video streaming application Fig. 5. Energy cost of each replica for distributed file service Fig. 6. Total energy cost of all replicas for distributed file service IV. CONCLUSION Our proposed system provides a decentralized architecture to solve the energy-aware replica selection problem for data- intensive applications in the cloud. It considers both total energy cost of the entire cloud and bandwidth capacity for each replica when forming the systemwide energy model. The performance of our prototype system proves that it is highly applicable to process different types of

algorithm used by our proposed system in order to achieve better performance and lower total energy cost. REFERENCES [1] the U.S. EPA ENERGY STAR Program. (2007) http://www.energystar.gov. [2] J. Dean and S. Ghemawat, “Mapreduce: simplified data processing on large clusters,” Commun. ACM, vol. 51, no. 1, pp. 107–113, Jan. 2008. [3] P. Wendell, J. W. Jiang, M. J. Freedman, and J. Rexford, “Donar: decentralized server selection for cloud services,” SIGCOMM Comput. Commun. Rev., vol. 41, pp. 231–242, August 2010. [4] A. Nedic, A. Ozdaglar, and P. Parrilo, “Constrained consensus and opti- mization in multi-agent networks,” Automatic Control, IEEE Transactions on, vol. 55, no. 4, pp. 922 –938, april 2010. [5] L. Barroso and U. Holzle, “The case for energy-proportional computing,” Computer, vol. 40, no. 12, pp. 33 –37, dec. 2007. [6] N. Kim, T. Austin, D. Baauw, T. Mudge, K. Flautner, J. Hu, M. Irwin, M. Kandemir, and V. Narayanan, “Leakage current: Moore’s law meets static power,” Computer, vol. 36, no. 12, pp. 68 – 75, dec. 2003. [7] B. Zhai, D. Blaauw, D. Sylvester, and K. Flautner, “Theoretical and practical limits of dynamic voltage scaling,” in Proceedings of the 41st annual Design Automation Conference, ser. DAC ’04. New York, NY, USA: ACM, 2004, pp. 868–873. [8] R. Masiero and G. Neglia, “Distributed subgradient methods for delay tolerant networks,” in INFOCOM, 2011 Proceedings IEEE, april 2011, pp. 261 – 265. [9] SystemG. (2008) http://www.cs.vt.edu/node/4666. [Online]. Available: http://www.cs.vt.edu/node/4666 506

Download (https://documents.tips/download/link/ieee2012-ieee-20th-international-symposiumon-modelling-analysis-simulation58aa27d7cb226) (https://documents.tips/documents/ieee-2012-ieee-20th-international-symposium-on-modelling-analysissimulation-58aa27d7cb226.html) Facebook

DESCRIPTION

Twitter

Google+

Email

More

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation-585f90b5d4212.html)

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems Towards Improving MapReduce Task Scheduling Using Online Simulation Based Predictions (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation-585f90b5d4212.html) Documents

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation-588e22616b498.html)

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems - A Novel Energy Efficient Broadcast Leader Election (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation-588e22616b498.html) Documents

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation-58815ed9c2666.html)

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems Dynamic Balanced Configuration of Multi-resources in Virtualized Clusters (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation-58815ed9c2666.html) Documents

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation-588cde83138bf.html)

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems Performance Comparison of Routing Protocols for Cognitive Radio Networks (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation-588cde83138bf.html) Documents

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation-58984031154ab.html)

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems Carbon-Aware Load Balancing for Geo-distributed Cloud Services (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation-58984031154ab.html) Documents

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-2013ieee-21st-international-symposium-on-modellinganalysis-simulation-589dae2161ecc.html)

[IEEE 2013 IEEE 21st International Symposium on Modelling, Analysis & Simulation of Computer and Telecommunication Systems (MASCOTS) - San Francisco, CA, USA (2013.08.14-2013.08.16)] 2013 IEEE 21st International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems - A Fixand-Relax Model for Heterogeneous LTE-Based Networks (https://documents.tips/documents/ieee2013-ieee-21st-internationalsymposium-on-modelling-analysissimulation-589dae2161ecc.html) Documents

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-200715th-international-symposium-on-modelinganalysis-and-simulation-589da26401e28.html)

[IEEE 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS) - Istanbul, Turkey (2007.10.24-2007.10.26)] 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems - Joint-State Modeling for Open Queuing Networks with Breakdowns, Repairs and Finite Buffers (https://documents.tips/documents/ieee2007-15th-international-symposiumon-modeling-analysis-and-simulation589da26401e28.html) Documents

(https://documents.tips/category/documents.html)

(https://documents.tips/documents/ieee-200715th-international-symposium-on-modelinganalysis-and-simulation-589e7d4f7adc7.html)

[IEEE 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS) - Istanbul, Turkey (2007.10.24-2007.10.26)] 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems - An Analytical Model for Torus Networks in the Presence of Hot-Spot Traffic with Batch Arrivals (https://documents.tips/documents/ieee2007-15th-international-symposiumon-modeling-analysis-and-simulation589e7d4f7adc7.html) Documents

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-200715th-international-symposium-on-modelinganalysis-and-simulation.html)

[IEEE 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS) - Istanbul, Turkey (2007.10.24-2007.10.26)] 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems - A Hierarchical Connected Dominating Set Based Clustering Algorithm for Mobile Ad Hoc Networks (https://documents.tips/documents/ieee2007-15th-international-symposiumon-modeling-analysis-andsimulation.html) Documents

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-200715th-international-symposium-on-modelinganalysis-and-simulation-589db66631f9f.html)

[IEEE 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS) - Istanbul, Turkey (2007.10.24-2007.10.26)] 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems - Model-free Adaptive Hysteresis for Dynamic Bandwidth Reservation (https://documents.tips/documents/ieee2007-15th-international-symposiumon-modeling-analysis-and-simulation589db66631f9f.html) Documents

(https://documents.tips/category/documents.html) (https://documents.tips/documents/ieee-200715th-international-symposium-on-modelinganalysis-and-simulation-58a10e6c81e99.html)

[IEEE 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS) - Istanbul, Turkey (2007.10.24-2007.10.26)] 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems - MDST: Multiprocessor DSP Simulation Toolkit for Voice Processing Applications (https://documents.tips/documents/ieee2007-15th-international-symposiumon-modeling-analysis-and-simulation58a10e6c81e99.html) Documents

(https://documents.tips/category/documents.html)

View more (https://documents.tips/search? q=%5BIEEE+2012+IEEE+20th+International+Symposium+on+Mod +Washington%2C+DC%2C+USA+%282012.08.72012.08.9%29%5D+2012+IEEE+20th+International+Symposium+on +Energy-Aware+Replica+Selection+for+DataIntensive+Services+in+Cloud)

TOP RELATED

(https://documents.tips/documents/ieee-2012-ieee-20th-international-symposium-on-modelling-analysis-simulation.html) (https://documents.tips/documents/ieee-2012-ieee-20th-international-sym (https

[IEEE 2012 IEEE 20TH INTERNATI…

212

0

COMPANY

[IEEE 2012 IEEE 20TH INTERNATI…

212

[IEEE 2012 IEEE 20TH INTERNATI…

0

CONTACT & LEGAL

212

0

OPENING HOURS

(https://facebook.com/d (https://twitter.com (https://goo

Smile Life

When life gives you a hundred reasons to cry, show life that you have a thousand reasons to smile

Get in touch

© Copyright 2015 - 2024 PDFFOX.COM - All rights reserved.