source: anr/section-2.2.tex @ 230

Last change on this file since 230 was 204, checked in by coach, 15 years ago

IA: version qui fait 2 page

File size: 13.2 KB
RevLine 
[25]1% Relevance of the proposal
[97]2The COACH proposal addresses directly the \emph{Embedded Systems} item of
[67]3the ARPEGE program. It aims to propose solutions to the societal/economical challenges by
[97]4providing SMEs novel design capabilities enabling them to increase their
[25]5design productivity with design exploration and synthesis methods that are placed on top
[97]6of the state-of-the-art methods.
7This project proposes an open-source framework for mapping multi-tasks software applications
8on Field Programmable Gate Array circuits (FPGA).
[99]9%%%
10\parlf
[97]11COACH will contribute to build an open development and run-time
12environment, including communication middleware and tools to support
[25]13developers in the production of embedded software, through all phases of the software lifecycle,
[177]14from requirements analysis downto deployment and maintenance.
[25]15More specifically, COACH focuses on:
16\begin{itemize}
17\item High level methods and concepts (esp. requirements and architectural level) for system
18design, development and integration, addressing complexity aspects and modularity.
19\item Open and modular development environments, enabling flexibility and extensibility by
20means of new or sector-specific tools and ensuring consistency and traceability along the
21development lifecycle.
22\item Light/agile methodologies and adaptive workflow providing a dynamic and adaptive
23environment, suitable for co-operative and distributed development.
24\end{itemize}
25COACH outcome will contribute to strengthen Europe's competitive position by developing
26technologies and methodologies for product development, focusing (in compliance with the
[204]27%scope of the above program) on technologies, engineering methodologies, novel tools,
28%methods which facilitate resource use efficiency. The approaches and tools to be developed
29%in COACH will enable new and emerging information technologies for the development,
30%methods which facilitate resource use efficiency. The COACH approaches and tools
31scope of the above program) on technologies, engineering methodologies, novel tools
32which facilitate resource use efficiency. The COACH approaches and tools
33will enable new and emerging information technologies for the development,
[25]34manufacturing and integration of devices and related software into end-products.
[99]35%%%
[191]36\parlf\noindent
[177]37The COACH project will benefit from a number of previous recent projects:
[99]38\begin{description}
39  \item[SOCLIB]
[204]40    The SoCLib ANR platform (2007-2009) is an open infrastructure
41    for system level virtual prototyping of shared memory, multi-processors
42    architectures. It provides tools to map multi-tasks software application on these
43%    The SoCLib ANR platform (2007-2009) is an open infrastructure developped by
44%    10 academic laboratories and and 6 industrial companies.
[199]45    (TIMA, LIP6, Lab-STICC, IRISA, ENST, CEA-LIST, CEA-LETI, CITI, INRIA-Futurs, LIS) and 6
46    industrial companies (Thales Communications, Thomson R\&D, STMicroelectronics, Silicomp, MDS, TurboConcept).
[204]47%    It supports system level virtual prototyping of shared memory, multi-processors
48%    architectures, and provides tools to map multi-tasks software application on these
49%    architectures, for reliable performance evaluation.
[99]50    The core of this platform is a library of SystemC simulation models for
51    general purpose IP cores such as processors, buses, networks, memories, IO controller.
52    The platform provides also embedded operating systems and software/hardware
53    communication middleware.
[204]54%    The synthesisable VHDL models of IPs are not part of the SoCLib platform, and
55%    this project enhances SoCLib by providing the synthesisable VHDL models required
56%    for FPGA synthesis.
[99]57    The synthesisable VHDL models of IPs are not part of the SoCLib platform, and
[204]58    this project enhances SoCLib by providing them.
59%  \item[ROMA] The ROMA ANR project (http://roma.irisa.fr, 2007-2010)
60%    involving IRISA (CAIRN team), LIRMM, CEA List THOMSON France R\&D,
61  \item[ROMA] The ROMA ANR project~\cite{roma}
[120]62    proposes to develop a reconfigurable processor, exhibiting high
63    silicon density and power efficiency, able to adapt its computing
64    structure to computation patterns that can be speed-up and/or
65    power efficient.  The ROMA project study a pipeline-based of
66    evolved low-power coarse grain reconfigurable operators to avoid
67    traditional overhead, in reconfigurable devices, related to the
68    interconnection network.  The project will borrow from the ROMA
69    ANR project and the ongoing joint INRIA-STMicro
70    Nano2012 project to adapt existing pattern extraction algorithms
[204]71    and datapath merging techniques to ASIP synthesis.
72%    and datapath merging techniques to the synthesis of customized
73%    ASIP processors.
[99]74  \item[TSAR]
[204]75%    The TSAR MEDEA+ project (2008-2010) involving BULL, THALES and the \upmc targets the design of a
76    The TSAR MEDEA+ project (2008-2010) targets the design of a
[99]77    scalable, coherent shared memory, multi-cores processor architecture, and uses the SoCLib
78    plaform for virtual prototyping. The COACH project will benefit from the synthesizable VHDL
79    models developped in the framework of TSAR (MIPS32 processor core, and RING interconnect).
80  \item[BioWic]
81    On the HPC application side, we also hope to benefit from the experience in
82    hardware acceleration of bioinformatic algorithms/workfows gathered by the
83    CAIRN group in the context of the ANR BioWic project (2009-2011), so as to
84    be able to validate the framework on real-life HPC applications.
85\end{description}
86%%%
[191]87\parlf\noindent
[97]88The laboratories involved in the COACH project have a well estabished expertise
[204]89%in the following domains:
90in the domains:
[97]91\begin{itemize}
[99]92  \item 
93    In the field of High Level Synthesis (HLS), the project
94    leverages on know-how acquired over the last 15 years with the GAUT~\cite{gaut08} project
95    developped by the \ubs laboratory, and with the UGH~\cite{ugh08} project developped
96    by the \upmc and \tima laboratories.
97  \item
98    Regarding system level architecture, the project is based on the know-how
99    acquired by the \upmc and \tima laboratories in the framework of various projects 
[135]100    in the field of communication architectures for shared memory multi-processors systems
101    (COSY~\cite{cosy}, DISYDENT~\cite{disydent05} or DSPIN~\cite{dspin08} of MEDEA-MESA).
102    As an example, the DSPIN project is now used in the TSAR project.
[99]103  \item
104    Regarding Application Specific Instruction Processor (ASIP) design, the
[120]105    CAIRN group at INRIA Rennes -- Bretagne Atlantique benefits from several years of
[99]106    expertise in the domain of retargetable compiler
107    (Armor/Calife~\cite{CODES99} since 1996, and the Gecos
108    compilers~\cite{ASAP05} since 2002).
[97]109\item
[99]110    In the field of compilers, the Compsys group was founded in 2002
111    by several senior researchers with experience in
112    high performance computing and automatic parallelization. They have been
113    among the initiators of the polyhedral model, a theory which serve to
114    unify many parallelism detection and exploitation techniques for regular
115    programs. It is expected that the techniques developped by Compsys for
116    parallelism detection, scheduling, process construction and memory management
117    will be very useful as a front-end for the a high-level synthesis tools.
[97]118\end{itemize}
[99]119%%%
[191]120\parlf\noindent
[177]121The COACH project answers to several of the challenges found in different axis of the
[191]122call for proposals.%Keywords of the call are indicated below in italic writing.
123\begin{description}
124\item[Axis 1] \textit{Architectures des syst\`{e}mes embarqu\'{e}s} \\
[177]125COACH will address new embedded systems architectures by allowing the design of
126Multi-Core Systems-on-Chip (possibly heterogeneous) on FPGA according to the design
127constraints and objectives (real-time, low-power). It will permit to design  complex SoC
128based on IP cores (memory, peripherals, network controllers, communication processors),
129running Embedded Software, as well as an Operating System with associated middleware and
130API and using hardware accelerator automatically generated. It will also permit to use
[191]131efficiently different dynamic system management techniques and re-configuration mechanisms.
132\textbf{Thereby COACH well corresponds to axis 1}.
133%
134\item[Axis 2] \textit{Infrastructures pour l'Internet, le calcul intensif ou les services} \\
[177]135COACH will address High-Performance Computing (HPC) by helping designer to accelerate an
136application running on a PC by migrating critical parts into a SoC implemented on an FPGA
137plugged to the PC bus (through a communication link like PCI/X). COACH will reduce the designer
138effort through the development of tools that translate high level language programs to FPGA
139configurations. Moreover, Dynamic Partial Reconfiguration will be used for improving HPC performance
140as well as reducing the required area.
[191]141\textbf{Thereby COACH partially corresponds to axis 2}.
142%
143% IA2PC: comme ce sont des axes tertiaire, il faut faire + court que primaire et
144% IA2PC: secondaire.
[204]145%VERS 3
146\item[Axis 3] \textit{Robotique et contr\^{o}le/commande} \\
147Manufacturing technology employs more and more SoC.
148COACH will permit to design such complex digital systems.
149\textbf{Thereby COACH indirectly answers to axis 3 too}.
150%\item[Axis 3 \& 5] \textit{Robotique et contr\^{o}le/commande} and \textit{S\'{e}curit\'{e} et suret\'{e}} \\
[191]151%VERS 1
152%Future control applications employ more and more SoC.
153%Application domains for such systems are for example the automotive domain, as well as the
154%aerospace and avionics domains.
155%In all cases, high performance and real time requirements are combined with
156%requirements to low power, low temperature, high dependability, and low cost.\\
157%Similary manufacturing, security and safety technologies require also more and more
158%computation power.
159%VERS 2 pour gagner de la place
[204]160%Manufacturing, controling, security and safety technologies employ more and more SoC.
161%COACH will permit to design such complex digital systems.
162%\textbf{Thereby COACH indirectly answers to axis 3 and 5 too}.
[191]163\end{description}
164%Axis 3 "Robotique et contr\^{o}le/commande}"
165%
166%COACH will permit to design complex digital systems based on high-performance multi-core systems.
167%Like in the consumer electronics domain (telecommunication, multimedia), future control applications
168%will employ more and more SoC not just for typical consumer functionality, but also for safety and
169%security applications (by performing complex analyses on data gathered with intelligent sensors,
170%by initiating appropriate responses to dangerous phenomena...). Application domains for such systems
171%are for example the automotive domain, as well as the aerospace and avionics domains (i.e. sophisticated on-board
172%radar systems, collision-detection, intelligent navigation...).
173%Manufacturing technology will also increasingly need high-end vision analysis and high-speed
174%robot control. In all cases, high performance and real time requirements are combined with
175%requirements to low power, low temperature, high dependability, and low cost.
176%
177%Axis 5 "S\'{e}curit\'{e} et suret\'{e}" :
178%
179%The results of the COACH project will help users to build cryptographic secure systems implemented in
180%hardware or both in software/ hardware in an effective way, substantially enhancing the
181%process productivity of the cryptographic algorithms hardware synthesis, improving the
182%quality and reducing the design time and the cost of synthesised cryptographic devices.
[177]183
[191]184% IA2PC: 1) je ne vois pas trop ce que ca fait la.
185% IA2PC: 2) c'est deja dans le 2.1 pour le small business.
186% IA2PC: 3) Pour le large business, on avait mis ca dans la premiere version et je pense
187% IA2PC     toujours que le large business est encore vise par COACH.
188% IA2PC     Alain a enleve toute reference sur ce large business. Sa raison est +
189% IA2PC     politico/stylistique: en parlant des 2 on n'est pas tres clair et on brouille
190% IA2PC     le message. Je partage assez son avis, la version actuelle est + claire que
191% IA2PC     celle d'avant. De plus on ne dit jamais que l'on ne vise pas les grosses
192% IA2PC     boites.
193% IA2PC
194% IA2PC Bref je serai assez pour enlever ce paragraphe, et ne pas faire reference au large
195% IA2PC business meme dans les section precedente. Par contre d'essayer de recaser le reste dans
196% IA2PC les sections precedentes.
197%
198% VERS 2 pour gagner de la place je l'enleve
199% COACH technologies can be used in both large and small business, as they will permit users to design
200% embedded systems which meet a wide range of requirements: from low cost and low power consuming
201% devices to very high speed devices, based on parallel computing. For enterprises that will use embedded
202% systems designed via the approaches and tools targeted by COACH, there is the potential for greater
203% efficiency, improved business processes and models. The net results: lower costs, faster response times,
204% better service, and higher revenue.
[204]205%\parlf
[177]206Finally, it is worth to note that this project covers priorities defined by the commission
[19]207experts in the field of Information Technolgies Society (IST) for Embedded
[199]208Systems: \textit{ $<<$Concepts, methods and tools for designing systems dealing with systems complexity
[19]209and allowing to apply efficiently applications and various products on embedded platforms,
[191]210considering resources constraints (delays, power, memory, etc.), security and quality
[199]211services$>>$}.
Note: See TracBrowser for help on using the repository browser.