Idea Transcript
“CAR AS A DATA CENTER” A TECHNO-ECONOMIC FEASIBILITY STUDY
THESIS SUBMITTED TO THE ELECTRONIC AND TELECOMMUNICATION DEPARTMENT OF POLITECNICO DI TORINO IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER IN WIRELESS SYSTEM AND RELATED TECHNOLOGIES
Yenny Marcela Olipa Buendía September 2013
I certify that I have read this thesis and that, in my opinion, it is fully adequate in scope and quality as a thesis for degree of Maste in Wireless System and Related Technologies
_____________________________ Giovanni Ghione (Master Coordinator)
I certify that I have read this thesis and that, in my opinion, it is fully adequate in scope and quality as a thesis for degree of Maste in Wireless System and Related Technologies
_____________________________ Antonio Manzanilli
i
ii
Objective Technological advances (e.g. standard h/w performance, embedded communications, device miniaturization, etc.) and the related costs reductions are progressively moving an incredible amount of processing, storage, communications-networking capabilities at the edge of traditional networks, i.e., in the hands of the end Users (and by Users it is meant not only people by also machines, smart objects, appliances and any device which is attached to the network at the edge). It is argued that this will transform the edge area into a fertile ground for the flouring of new ICT ecosystems and business models.
Main objective of this work is to define, model and analyze how developing a ICT service ecosystem whose “seed” is the concept of the “Car as a Data Center”. The use case will look at how exploiting the most recent advance in ICT technologies (e.g., Cloud and Fog Computing, pervasive Software Networks, etc), to transform vehicles in network service nodes. This will be achieved both by embedding into cars processing, storage and communications resources and creating “virtual images” of the vehicles on a Cloud Computing infrastructure in this sense the virtual car on the Cloud could be seen also as a gate to access any ICT services (e.g. even provided by the Car Producer, or any Third Parties) or any data associated to the car.
iii
Abstract Technology is undergoing impressive progress in areas as processing, storage and communications. Processing is continuing to follow Moore’s curve in order to increase its capabilities which is doubling in capability roughly every 18 months. Store capacity on a single chip is doubling roughly every 12 months. The communications area is undergoing progress with the fiber optic. The optical bandwidth is doubling every 9 months. The performance of processing and storage technologies are making impressive progress, but future networks will rely on software. In fact, currently advances in resources virtualizacion are allowing the diverse coexisting of virtual network on the same physical infraestructure. Software Defined Networks (SDN): allows network administration to manage service through abstraction of lower level funcionality. It might be confused with network virtualization because sometimes both trajectories could intersect. However, sometimes network virtualization allows the set up virtual networks by connecting virtual IT and networking resources. In our project we want to take advantage of the recent advances in processing, storage and communication by defining a new ICT services ecosystem around vehicle. It will allow us to interact with the enviroment through the plenty of application that will be installed in a sort of black box inside the car. Already today there is a similar project which is being developed by Google. The robot vehicle is called Stanley and its system is driven at the speed limit it has stored in its maps and maintains its distance from other vehicles using its system of sensors. The system provides an override that allows a human driver to take control of the car by stepping on the brake or turning the wheel. The vehicle have completed over 300.000 autonomous driving miles accident free and the team are starting to test it with single driver instead of pair. Google has not even plans to commercially develop the system because they hope to develop a business which would market the system and the data behind it to automobile manufacture. They have also realized that the technology is ahead of law in many areas. Thus, it is iv
necessary that a law is passed for allowing autonomous motor for starting to commercialize this system. Against this background, we are going to define a model and develop a new ICT services ecosystem by exploiting the most recent advances in ICT technologies. A definition of natural ecosystem and a briefly description of the five steps of the development and analysis of an environment will be described in the first part of our project. After that, we are going to describe in detail the five steps of ecosystem modeling. The first step is to define a business ecosystem for our project based on natural ecosystem definition. The second step is the identification and collection of the relevant technologies and elements of the our ecosystem where we will identify the relevant technologies e.g. connectivity and cloud computing. The third step is to identify the players in our ecosystem in order to describe each of them (e.g. Product manufacturer, network provider and so on). In this step is important to build a connection matrix which allow us to know how strong is the relationship among the players. The fourth step is to make an ecosystem analysis which will include the identification of criteria to evaluate the ecosystem’s behaviors and relevant key indicators. The last step is about requirements and specifications, where we will describe the specifications of a real use case. Finally, we will include conclusion, recommendations and proposals.
v
vi
Contents 1. Contents Abstract……………………………………………………………………………………………………………………….……………………iv 1.
THE BUSINESS ECOSYSTEM ............................................................................................................. 1
2.
ECOSYSTEM MODELING .................................................................................................................. 2 2.1
STEP 1: DEFINITION OF ECOSYSTEM ....................................................................................... 2
2.2 STEP 2: IDENTIFICATION AND COLLECTION OF THE RELEVANT TECHNOLOGIES AND ELEMENTS OF THE ECOSYSTEM. ......................................................................................................... 3 2.1.1
Relevant Technologies .................................................................................................... 3
2.1.2
Players ............................................................................................................................. 3
2.3
STEP 3: ECOSYSTEM MODEL REPRESENTATION..................................................................... 6
2.4
STEP 4: ECOSYSTEM ANALYSIS ................................................................................................ 8
2.4.1
Number of Users ............................................................................................................. 8
2.4.2
Applications ................................................................................................................... 11
2.4.3
Market value for the car business ................................................................................. 15
2.5
STEP 5 USE-CASE REQUIREMENTS AND SPECIFICATIONS .................................................... 16
2.5.1 3.
Technical Feasibility ...................................................................................................... 18
Conlusion and Recommendations ................................................................................................ 23
List of references
vii
List of Tables Table 1 Connection Matrix ...................................................................................................................... 5 Table 2 Number of licensed cars in Ecuador 1997-2011......................................................................... 9 Table 3 The most sold brands in Ecuador ............................................................................................. 10
viii
List of Figures Figure 1 Biz Ecosystem ............................................................................................................................ 2 Figure 2 Virtual Images on a Cloud Computing Infrastructure ............................................................... 6 Figure 3 Relationship among players ...................................................................................................... 8 Figure 4 Global ARPU for LTE subscribers ............................................................................................. 11 Figure 5 Self-driving car......................................................................................................................... 15 Figure 6 The open Source Cloud Operating System.............................................................................. 21
ix
x
1.
THE BUSINESS ECOSYSTEM
A business ecosystem is approached by examining a biological ecosystem, how they involve and how they are classified and structured. Thus, first we are giving a definition of a natural ecosystem which is defined as a biological community of interacting species plus their physical environment. It varies greatly in size and the elements that makes it up, but each is a functioning unit of nature. Everything that lives in an ecosystem is dependent on the other species and elements that are also part that ecological community. If one part of an ecosystem is damaged or disappears, it has an impact on everything else. Business ecosystem is a relatively new concept in the field of business research but we can define the business ecosystem how “the network of buyers, suppliers and makers of products or services” plus the socio-economic environment, including the regulatory framework. To describe the ecosystem we are going to use a methodology described in five steps which allows us to develop and analyze the ecosystem. The five steps of development and analysis are: a) Definition of ecosystem: The first step is identify the seed, something that attracts actors thinking it could be the leverage to generate a business, then define the business ecosystem around our seed.
b) Identification and collection of the relevant technologies and elements of the ecosystem: Identify the list of classes of information to be collected to the constitutive elements and relationship
c) Ecosystem model representation: To develop a representation model of the ecosystem able to include all remarkable features characterizing its behaviors.
d) Ecosystem Analysis: Identification of criteria to evaluate the ecosystem’s behaviors and relevant key indicators. 1
e) Use-case: Requirements and specification: Identification and description that are developing from the data collected in the considered temporary frame.
Figure 1 Biz Ecosystem
2.
2.1
ECOSYSTEM MODELING
STEP 1: DEFINITION OF ECOSYSTEM
The basic idea or “seed” of the ecosystem is anything around us, this will act as a gate to access data and service ICT (Information and Communications technologies). Moreover this will have a regulation framework. We can define our ecosystem as a network of any device around us, technology suppliers, Telco providers, Data providers , customers and regulatory framework which allow us to access data and services ICT.
2
2.2
STEP 2: IDENTIFICATION AND COLLECTION OF THE RELEVANT TECHNOLOGIES AND ELEMENTS OF THE ECOSYSTEM.
2.1.1
Relevant Technologies •
Devices and consumers electronics: This group represents all the devices that surround us, within this group is involved devices such as camera, mobile phone, television, refrigerator and so on.
•
Connectivity: The connectivity could be fixed or mobile. Fixed connectivity is when we access data and services ICT through Wi-Fi connection which is part of a LAN (Local access network) network with a wireless modem. Instead, mobile connectivity is given by a Telecom provider. Thus, we can define by connectivity as the path to access data and information.
•
Cloud: It represents the virtual place where the data will be either processed or stored.
•
Network and Service platforms: It represents all the devices and services behind the Telecom providers, which allows us to access data and information.
2.1.2
Players • Users: The users will be able to access data and information through any device or product around them. •
Consumers’ electronics and device providers: In this group are involved all the manufacturer companies of electronics devices (e.g. camera, mobile phone, television, and so on). For instances: Nokia, Samsung, Apple, etc.
•
Product manufacturer: This group is referred to products as glasses, tables, desks, chairs, etc. That’s means all the products around the customers which 3
are not electronics devices but have either a circuit embedded or a bar code whose allow us to enter a web page using a mobile phone. In this web page we will find information of the product and extra information related to the product. • Application Developers: This type of actors are who develop programs for use in business, they are always developing new types of applications for computers, mobile phones and other type of electronics devices. (e.g. Apple, Google, etc.) • Networks Providers: Networks providers are organizations that
sells us
bandwidth or network access by providing direct Internet backbone access to the Internet. Network service providers may consist of telecommunication companies, data carriers, wireless communications providers and cable television operators offering high speed Internet access. (e.g. Telecom Italia, Orange, FastWeb, etc.) • Over the Top: This term is used for calling to the big companies which offer storage and processing services in the “cloud” such as Google, Facebook, Microsoft, YouTube, etc. • Equipment Providers: are companies that provide communication solutions to Network Providers as fixed or mobile operator. The appliances are part of the access and core network. For instance: Nokia, Huawei, Siemens, Cisco, etc. •
Network software developers and integrators: are companies that provide software and integration solution for use in business. They are always developing news software which main achieve is to integrate proprietary interfaces in just one. Once accomplished this aim, it is more easy work with just one interface and the network providers avoid working with only one equipment providers. For instance : NEC, Microsoft, Cisco, Facebook, Google, etc. 4
• Service Providers: A service provider is any organization that provides services such as consulting, education, communications, storage, processing and many other services to the public. IT professionals sometimes differentiate between services providers by categorizing them as types: internal service provider, shared service provider, external service provider. (e.g. Telecom Italia, Wind). • Information providers: are companies that provide data, knowledge and information to customers and collects it from its contributors to manage and store it by optionally using administrators. • Semiconductors Providers: The companies that are part of this group offers elements such as chipsets, FET and so forth. These are one of the main materials for making the appliances work. This devices are part of a base
2 2 1 1 3 1 2 0 0 0 0
1 1 2 0 0 1 0 0 0
1 2 0 0 1 1 0 1
1 0 1 2 2 0 0
3 2 2 1 1 2 2 2 0 0
1 0 0 0 1 1 1 0 1 0
2 0 0 1 2 1 1 0 0 2
1 1 1 2 2 1 1 2 2 0
0 0 1 2 2 0 0 2 2 0
0 0 0 0 0 1 1 1 2
Semiconductor Providers
Information Providers
Services Providers
Network Software Developers and Int.
Equipment Providers
1 1 1
Over top
1 1
Network Providers
Application Developers
User Consumers' Electronics and Device Provides Product Manufacturer Application Developers Network Providers Over the top Equipment Providers Network Software Developers and Int. Services Providers Information Providers Semiconductors Providers
Product Manufacturer
User
0 - No relationship is On 1 - Intangible relation 2 - Tangible Relation 3 - Possible Future Relation
Consumer's Electronics and Device Provides
station or core network.. (e.g. INTEL, AMD).
0 0 1 0 0 0 2 0 0 0
0
Table 1 Connection Matrix
5
2.3
STEP 3: ECOSYSTEM MODEL REPRESENTATION
In this step we will develop a real ecosystem model, where we will describe each component and remarkable features. For our project the object or seed of the ecosystem will be the concept of “Car as a Data Center”, this will be attained both by embedding into cars processing , storage and communications resources and creating a “virtual images” on the Cloud
LTE connection
Cloud Computing
User Figure 2 Virtual Images on a Cloud Computing Infrastructure
The components of this ecosystem are: •
User: is anyone who is interested in accessing any ICT services from their vehicles having a safe and entertaining trip without wasting time for instance, looking for spots for parking.
•
Cloud Computing: Cloud Computing framework allow us to create a “virtual images” of the vehicle which could be seen as a gate to access any ICT services. Once the data will be gathered by sensors, the data will be processed and stored on a Cloud, because it maximizes the effectiveness of the shared resources by multiple users. 6
Furthermore the maintenance of cloud computing applications is easier , because they do not need to be installed on each user’s device and can be accessed from different places. Behind cloud computing are Car Producer, any Third Parties, also the well know “over the top” companies as Google. In addition, the data collected through the equipment installed inside the vehicle could be important by third parties in order to analyze and predict behaviors and tends. •
LTE Connection: The way of connectivity between the embedded resources into cars with Cloud where a “virtual images” will be create could be possible trough LTE technology due to bandwidth that user can experience. The players behind LTE Connection are both Telecom Operators (e.g. Telecom Italia, Claro Ecuador) and the Equipment Providers (e.g. Huawei, Nokia).
•
Network and Services platforms: In order to access to the data storage in the Cloud it is necessary to have a platform which allow sharing data. For our example the players behind the network and services platforms are Telecom Operators and third parties.
•
The object: “Car as a data Center” will be the object of our project t, It will be equipped with processing and storage appliances in order to gather and analyze the data took from sensors installed along road and into cars.. This allow to the user get fresh and real information about car’s conditions as optimum oil level, tires performance, engine and filters states. Regarding environment information, the user can be able to know whether forecast, traffic jam, best road and so on. Furthermore, the data collected will be send to the cloud thorough a LTE technology. For our example the players behind the object are car manufacturers such as BMW, Fiat, etc; tech providers such as third parties, Network providers such as Telecom operators and Application Developers.
7
IT Providers
User
Car Manufacter
Telecom Providers
Tech Providers Data End
App. Providers Figure 3 Relationship among players
2.4
STEP 4: ECOSYSTEM ANALYSIS
This step is devoted to analyze the ecosystem in a real example, the environment that we have chosen is Ecuador because it is my home country as well as we would like to analyze how feasible would be implement this project by using it as example of developing in our region.
2.4.1
Number of Users
Firstly, we have to define how big is our environment. The last census made in 2010 showed that Ecuador’s population is 14’306.846 people 1.
1
INEC Instituto Nacional de Estadísticas y Censos. Last census 2010.
8
According to the INEC (Instituto Nacional de Estadísticas y Censos), the number of licenced cars during 2011 was 1.418.339. The figure 2 depicted how has incresed the number of vehicles from 1997 to 2011. It is also important to add that according The World Bank, there are 71 vechicles per each 1000 people according to the official data released in 20122..
VEHICLE USE YEAR 1.997 1.998 1.999 2.000 2.001 2.002 2.003 2.004 2.005 2.006 2.007 2.008 2.009 2.010 2.011
TOTAL
PERSONAL
HIRE
STATE
LOCAL GOVERNMENT
561.864
531.189
22.842
6.184
1.649
587.350
554.040
25.611
5.959
1.740
624.924
592.252
25.700
5.284
1.688
646.040
617.116
23.047
4.481
1.396
621.181
594.206
20.503
4.882
1.590
663.231
624.466
32.176
4.786
1.803
723.176
679.548
34.949
6.712
1.967
764.086
726.867
29.691
5.225
2.303
867.666
827.166
30.504
7.530
2.466
961.556
915.089
38.644
5.613
2.210
920.197
873.697
36.959
6.929
2.612
989.039
953.199
29.590
4.773
1.477 1.565
905.651
867.387
29.370
7.329
1.171.924
1.116.201
37.352
14.567
3.804
1.418.339
1.354.346
45.282
14.575
4.136
Table 2 Number of licensed cars in Ecuador 1997-2011
3
Now, we are going to focus in the amount of embedded processors exists into a car, for instance in a current 7-Series BMW and S-class Mercedes boast about 100 processors apiece. A relatively low-profile Volvo still has 50 to 60 baby processors on board. Even a boring low-cost econobox has a few dozen different microprocessors in it. With this data we can figure out that there are more that hundres of embedded processor working on google selfdriving car gathers huge amounts of data in order to make a decision while driving. During 20114, the most sold vehicle’s brands
was Chevrolet and
Suziki which
are
subsidiary of General Motors, another brands are depicted in figure 3. We can noticed that most of Ecuadorian s are afforded to buy a vehicle between $14.000 and $40.000.
2
The World Bank [http://datos.bancomundial.org/indicador/IS.VEH.NVEH.P3] INEC ECU-INEC-DIEC-EDT-2012-v1.8 4 Ekos Negocios, release 2011 3
9
Number of automobile (December, 2011) Chevrolet + Suzuki 4.860 Hyundai 1.207 Nissan 855 Toyota 646 Mazda 644 Kia 602 Ford 510 Great Wall 217 Renault 200 Chery 178 Volkswagen 156 Skoda 56 Mitsubishi 56 Lifan 25 Table 3 The most sold brands in Ecuador Brand
Another important economic indicator is ARPU (Average Revenue per User) which was up 6.5 percent to $10 relative to the prior year (2011)5 data taken from America Movil’s subsidiary Claro Ecuador which has more than 8.5 millons of subscribers turn it into the operator with more amount of subscribers. We have included the virtual world games with lots of refreshing’s DARPU (diarily average revenue per unit), which is near $0.05 and We have also included Facebook’s earned an average which is $1.28 revenue per user regarding the official data released in the second quarter of 20126 Telecommunications ARPU is expected to further decline over the next few years regarding voice service but the demand for bandwidth will increase exponentially. According to this backgroung Mobile networks operators must look to find innovate ways to reduce costs and monetize their network assets.
5
América Móvil http://www.insidefacebook.com/2012/07/27/facebook-average-revenue-per-user-was-1-28-in-q2-but-stillbelow-q4-2011-average/ 6
10
Figure 4 Global ARPU for LTE subscribers
7
How we can observe from the figure 4, the ARPU for connectivity is falling each year around the world for this reason the development of new application will be the new way of make business for mobile network operators.
2.4.2
Applications
According to this background, we propose to develop some applications to our environment: Car Functioning applications
7
•
Check the pressure of air
•
Measure of oil
•
Check of spark plug
http://www.intel.com
11
•
Automatic control of temperature
•
Check of battery life
•
Adjust volume in relation to read noise
•
Check of engine
These applications will allow us have a safe trip with applications as check the pressure of air or check of engine. If they detected something wrong, they send us warnings in order avoid accidents or damages on the car. This applications will be possible developed using the sensors. These devices will be installed into the car which will be gathering information in order to send the data to the processing and store black box.
Car Navigation applications •
Surf the Internet
•
Traffic
•
Road – Maps
•
Parking
•
Auto Drive
•
Weather Forecast
•
GPS location
These applications allow us to optimize time either for looking for the best road to reach our destination without traffic jam or to look for parking. Furthermore allow us to look up the weather forecast, receive mails, check the facebook and so on because we will be connected to Internet. For instance, parking application will tell us if there are some availables and legal spots for parking and how long we can stay there. We can add to powerful features on this applicacion, por example: while the vehicle are reaching the spot, a “conversation” between the vehicle and parking machine can establish previously in order to avoid getting the ticket.
12
Infotainment applications •
Entertainment (video and audio streaming)
•
Information (news)
•
Advertisement
•
Multiplayer games
These applications allow us to have a comfortable and relaxed trip as well as being informed all the time by receiving selected news. Finally advertisements will be presented via Google’s search engine because the emergence of this form of advertising is growing. In order to attain a good perfomance of these application, enough bandwidth is necessary to maintain a good user experiencie.
The following application are focus on process orientation: Processing and Networking power •
Connectivity
•
Car as a node
•
Car as a route
In order to ensure that data is gathered, processed, stored, shared and also available all the time we must ensure connectivity from sensors to box black to cloud without failures turning the car into either a node or a route. Most of these applications are either feasible today or in few years. However a feature application will be auto drive. Driverless car is a project which has as objectives reduce accidents , reduce time and energy wasted in traffic and reduce the number of cars on road, that’s means it is capable of fulfilling the human transportation capabilities of a traditional car. Taking into account that human error accounts for between 77 and 90% of all road accidents, it is really a big percent so Google’s approach is to get drivers out of the loop rather than make them better drivers. 13
To turn this challenge into something real are working twelve well know engineer and the company has spent around $50 million on this project which is less than 0.0003 percent of Google’s revenue. Google business would be to supply driverless software to manufacturers, rather than making cars itself, it might warrant Google’s perusing the Android strategy again by licensing the software for a price so low than manufacturers cannot refuse. If this software works Google could provide to millions of cars map and traffic data to Google maps at the same time Google would be feeding location and behaviors data from customers allow it to exchange queries and advertising via Google’s search engine. But how will this software operate? A million of sensor will have to be installed along the road and into the cars as well, it will also need cameras, array of radar and laser-based range finder which allow self-driving and provide us applications such as traffic, forecast, entertainment and so on. From the figure 5 which shows us how a car would see while it is driving and about to make a left turn, the car is capturing every single thing that it seem to moving in order to make a decision. It will be huge the amount of data that it would have to analyze as the movement of birds, rolling balls and another cars all that together in less than a second before making a decision, processing capabilities is also present in this project as well as advanced control systems which will interpret sensory information to identify appropriate navigation paths, obstacles and relevant signage.
14
Figure 5 Self-driving car
8
But not only will need of sensors to gather data, the driver less car will be able to sense their surroundings with such a techniques as radar and GPS in order to update their maps based on sensory input allowing the vehicles to keep track of their position even when conditions change but before that will be necessary to gather data about environments and then add it to highly detailed maps of the roads and terrain in order to be compared with the data it is acquiring from all those sensors and cameras to the previously recorder data. In spite of the big progress there are a lot challenges to overcome as the driverless car will be capable to drive over heavy rain or snow road.
2.4.3
Market value for the car business
This section is devoted to estimate the price that would have the device on the market. In order to make an analyze, we are going to apply the concept of contiguous markets
8
http://www.kurzweilai.net
15
Firstly we are going to analyze how much people are willing to pay by a car insurance: Ecuadorians are willing to pay around $18009 (i.e. depend on model and how new is the car). They are also willing to pay about 42,27% from the minimum wage for trips, leading technologies goods and entertainment. Moving from others areas to our ecosystem we figure out that Ecuadorians will be willing to pay by new applications because it is represented less amount of money, for instance if we compare how much they have to pay for an accident insurance which is hundreds of dollars with how much they would have to pay by an auto driver application which will much less for sure, the profit will be high, because It is not only to save money it is also to avoid accidents and people injuries. Furthermore, we will offer three types of appliance with different applications from the most basic installed in all devices to the most powerful applications installed in luxury vehicles, depending on how much someone will afford to pay for get the applications
2.5
STEP 5 USE-CASE REQUIREMENTS AND SPECIFICATIONS
This step aims at defining the requirements and the specifications for developing a use-case, which a car as a data center. Requirements: •
Black box needs capabilities in: Processing Storage Communication Networking Software
9
•
Application to handle data
•
Hall effect sensor
Seguros Equinoccial Ecuador
16
•
Parking sensors
•
Radar gun sensors
•
Speedometer sensor
•
Speed sensor
•
Throttle position sensor
•
Tire-pressure monitoring sensor
•
Transmission fluid temperature
•
Variable reluctance sensor
•
Vehicle speed sensor
•
Wheel speed sensor
•
Humistor sensor
•
Rain sensor
•
Open source platform
The main features of the device will be virtualization of the base station platform to attain the same benefits that they attain in their data centers, ensure processing capabilities due to the stringent real-time requirements of processing ratio , provide capabilities of storage in a centralized database. Sensor's capabilities are also important, there are innumerable applications for sensors of which most people are never aware. Nowadays, we can find capabilities of processing and memory into a sensor but the problem that has been present all the time is to save energy, there are some approaches that are looking for improve battery life, approaches have left the concept "always on" by hibernate and deep sleep mode for instance. For our project we are looking for sensors which have features as saving energy, some processing capabilities (e.g routing) and storage capacity because they will be deployed along the road and we need that they will be able to maintain the network.
17
2.5.1
Technical Feasibility
To develop this project already exists the necessary technology to come it true. It needs connectivity, processing and software capabilities. Each of them we are going to explain in details:
Connectivity In order to ensure connectivity we propose to use either mobile network or WiFi network.
Mobile Network: Historically, mobile network operators have competed based on providing the best coverage, best price and best voice quality. Nowadays, with the transition to multimedia services the mobile network operator are starting to compete based on the highest data rate because the demand for bandwidth has been growing year on year, driven by faster computers, higher resolution displays and the tendency to download more images and videos. Faced with an exponentially increasing demand for bandwidth and an ARPU further decline over the few years, mobile operator network are looking for innovative ways to reduce costs on implementation. New architecture approaches as femtocells, relay stations and closer interworking with Wi-Fi are being replaced for the migration to a mobile Software Defined Network (SDN) enabled by virtualization technologies. SDN approach allows network administrators to manage network services through abstraction of lower level functionality. A configuration of SDN can create a logical network control plane where hardware is physically decoupled from the data forwarding plane hardware, thus the control plane could be implemented using a different distribution model than the data plane Mobile network operators expects to provide a good user experience encompasses both data rate and reduced round-trip-time (RTT). Making use to SDN approach, it allows to move 18
closer to subscribers, all the way to the network edge and on the base station. Moving network services, such a DNS and HTTP, to the network edge reduces the RTT associated with the protocol handshakes. Virtualization of base station platform include benefits as cost reduction, ability to rapidly rollout new service offering and the ability to monetize edge network services. Currently, in Ecuador there are 3G and 3.5G technology implemented, the last technology wide implemented was HSPDA+ which reach speeds of 21 and 336 Mbps allowing the deployment of new applications. LTE technology is being implemented since some months ago in the main cities: Quito and Guayaquil by CNT (Coorporación Nacional de Telecomunicaciones) after that, CNT will start to implement it in the rest of the country. The speed of LTE will be ten times HSPDA+' speed. To keep connectivity from the vehicle will be necessary to have an SIM chip embedded which allow the vehicle to keep connecting to the mobile network. This represents an extra cost for maintain connectivity.
Wi-Fi Connectivity We are interested in using this technology because we want to create an Mobile Ad-hoc Network (MANETs) among black boxes (i.e hardware appliance installed into the vehicle) smart phones and sensors which will be called nodes. The network will be maintained partially by the nodes because they rely on some infrastructure due to gateway nodes which allow us to surf the Internet and use all the applications. The implementation of this technology allow us to maintain connectivity among cars and the data flow will be constantly proceeded allow us to share this information in order to maintain upgrade the network. But we also need communication network interacting with navigation system to find alternative routes. Devices for adaptive traffic control (e.g. smart traffic light) To make much faster the time of response, the smartphone will become a portable Wi-Fi hotspot and a IT sensor which allow us to forward traffic information car to car. 19
Using a smart phone (in a car) as a Wi-Fi route for providing communication capabilities as well as we can take advantage of this technology and have a network for free.
Processing Processing capabilities will be coverage by pizza box server, this term refers to the shape of a computer server enclosed in a rectangular and horizontally arranged chassis, the server does not need a lot of horsepower, has lower specifications silicon and thus lower cooling requirements. Because of this it can be reasonably small, around 19” wide and about 20” deep. In any country where Italian pizza can be ordered for home or office delivery, it comes in a box of roughly the same size as one of these servers. Using a pizza box server for providing the processing capabilities we ensure processing because this server can handle server to storage traffic.
Software Software capabilities will be coverage by Open Stack which is a global collaboration of developers and cloud computing technologists producing the ubiquitous open source cloud computing platform for public and private clouds. Open Stack is a cloud operating system that controls large pools of compute, storage and networking resources throughout a datacenter all managed through a dashboard that gives administrators control while empowering their users to provision resources through a web interface, the figure 6 depicts how works open stack operation system. From the figure 6, we can observe that open stack share services as Compute, Networking and Storage. The term compute refers to allow enterprises and services providers to offer ondemand computing resources by provisioning and managing large networks of virtual machines. The compute architecture is designed to scale horizontally simple by adding news servers on standard hardware providing flexibility with no proprietary hardware or software requirements, that mean if some server fails OpenStack replicates its content from other active node to new locations in the cluster. It is possible because OpensStack use software 20
logic to ensure data replication and distribution across different devices avoiding to spent in expensive equipment. Compute resources are accessible via APIs for developers building cloud application and via web interfaces for administrators and users. The requirements for storage had been changed, nowadays the enterprises choose them depending on their needs in performance and price requirements. OpenStack has support both object storage and block storage. For our project we have decided to choose object storage because it provides a fully distributed API-accessible storage platform that can be integrated directly into applications, offers a distribute storage system for static data such as virtual machine images, photo storage, email storage, backups and archives. To achieve this, objects and files are written to multiple disk drives spread throughout servers in the data center, but under the responsibility of it for ensuring data replication and integrity across the cluster. Furthermore, having no master point of control provides greater scalability, redundancy and durability.
10
Figure 6 The open Source Cloud Operating System
10
www.openstack.org
21
Instead that Block Storage allows block devices to be exposed and connected to compute instances for expanded storage, better performance and integration with enterprise storage platforms.
Regarding networking it can be used by administrators and users to increase the value of existing datacenter assets. OpenStack Networking ensures the network will not be the bottleneck or limiting factor in a cloud deployment and gives users real self-service. Another important feature is that it manages IP addresses, allowing traffic to be dynamically rerouted to any compute resource in case of failure. Furthermore, It provides flexibility networking models to suit the needs of different applications or user groups and Administrators can take advantage of software defined networking (SDN) technology to allow for high levels of multi-tenancy and massive scale. Through dashboard which is a graphical interface, administrators can control their compute, storage and networking resources in other words access cloud-based resources. It offers an overall view of the size and state of your cloud allow administrator creates users and projects, assign users to projects and set limits on the resources for those projects. Some advantages to use Open Stacks are: •
Simple to implement
•
Flexible enough allow us to build up our project however we want it
•
Have full control over hardware
•
Ability to innovation on the cloud by providing additional solutions
•
Stable platform which give us tools and skills to deploy applications.
•
Having not price, it is low cost allow the enterprise do not charge over customer for any license.
22
3.
Conlusion and Recommendations
•
This project has been modeled and analyzed the potential development of a business ecosystem based on the concept of “Car as a Data Center”. We described each step of ecosystem modeling in detail which allowed us to give a clear definition about our business ecosystem as well as to identify and gather information about the most relevant technologies and elements which allowed us to build up a connection matrix where we identified the different relationship among the players The next step was to make an ecosystem model representation where described each component and remarkable features of each component of our ecosystem. At ecosystem analysis’s step we described the amount of user of our environment and give some economical measure to determine if the user are willing t par by the applications, we also developed a technical feasibly section where described the capabilities that should be our device which we defined as black box. Finally we defined the requirements and specifications for developing a use-case.
•
To develop this business and technical project we had researched and learnt about how defined a business environment, how to measure the revenues in business as telecommunications for finding contiguous markets as well as we had research about Google’s car, OpenStack, sensors, pizza box server, android platform and the development of microprocessor and communication technologies.
•
From the develop of our project we have realized that it is necessary to acquire or improve abilities in software developing because the technologies is migration toward using open source because It offers us low cost as well as there is a global software community of developers collaborating on a standard and massively scalable open source cloud operating system. This enforces us to innovate in our products in order to be part of the market.
23
•
To be affordable our project it is necessary to focus on free platforms, develop MANET networks and offer innovation applications, all of that is necessary because the cost will reduce if we use free platforms and free connection as much as possible allowing customers be willing to pay.
•
Our purpose by developing this project has been to know what we need to run this project in Ecuador, economical and technical features has been researched in order to know how feasibly it is.
•
After developing of this project we concluded that it is feasible to implement this project in Ecuador, most of the technologies as connectivity, processing and software are already implemented and the others is been implemented. However auto drive application needs more research and appliances that results still expensive for implementation.
•
The subscriber’s expectation is that higher data rates will provide the best user experience for this reason mobile network operators is redefined their business model in order to become an essential part of the Internet applications and services of the over-the-top (OTT) revenue stream.
•
Among the new technologies for improve the Radio Access Network, SDN architecture approach is a greatest promise because enables by virtualization technologies. This approach enables virtualization of base station, services that were in the data center are moving to the network edge allowing mobile network operators attains cost saving, performance advantages, significantly improve the user experience, and drive new revenue allowing mobile network operator maximize their ARPU to offset the increase in CAPEX (Capital Expenditure) and OPEX (Operational expenditure) associated with network capacity expansion.
24
•
Regarding virtualization of hardware and computational resources has underpinned the advances in data center and cloud computing allow us to work over platforms such as OpenStack to deploy the cloud without to share any cost.
•
Regarding driverless technology there are two forces which are looking for developing this new approach and introducing it on the market, however Google is thinking fast in an approach which gets drivers out of the loop, car market’s perspective is incremental approaches in order to introduce this new technology bit by bit But whether Google achieves first the aim, Google would sell the software at low cost forcing to car market’s to be part of Google’s customers.
25
List of Reference •
Mirva Peltoniemi; Elisa Vuori: Business Ecosystem as the new approach to complex adaptive business environments. 13p. [http://iwoce.org/definitions-of-ecosystems.pdf]
•
INEC
Instituo
Ecuatoriano
de
Estadísticas
y
Censos.
Last
Cence
2010,
2010,
pg1[http://www.eclac.cl/celade/noticias/paginas/3/45123/ecuador_afiche.pdf] •
Asociación
de
Empresas
Automotrices
del
Ecuador
AEADE
[http://www.aeade.net/web/index.php?option=com_content&view=article&id=145&Itemid=80] • • • • • • •
INEC Instituto Ecuatoriano de Estadísticas y Censos, ECU-INEC-DIEC-EDT-2012-v1.8 [http://anda.inec.gob.ec/anda/index.php/catalog/174] America Movil pdf p.12 [ http://www.americamovil.com/amx/es/cm/reports/Q/2011_1.pdf] Dave Smith, Jeff Friesen: Android Recipes Problem-Solution Approach, 2nd Edition Nick McKeown, Software-defined Networking , Infocom, April 2009 p [http://www.cs.rutgers.edu/~badri/552dir/papers/intro/nick09.pdf] Ken Pepple, Creating Open Source Clouds, Developing Open Stack, July 2011 Kris Jamsa, Computing, SaaS, PaaS,IaaS, Virtualization, Business Model, Security and more, 2013 David J. Anderson, Agile Management for Software Engineering, 2004 p150.
26