source: anr/section-2.2.tex @ 153

Last change on this file since 153 was 135, checked in by coach, 15 years ago

IA: fixed references

File size: 8.2 KB
RevLine 
[25]1
2% Relevance of the proposal
3
[101]4\mustbecompleted {FIXME == AJOUTER LES POINTS QUI SUIVENT sur les 3 axes ???}
5
6--------------------------------------------------------------------------
7The COACH project answers to several of the challenges found in different axis of the call for proposals. Keywords of the call are indicated below in italic writing.
8
[134]9Axis 1 "Architectures des syst\`{e}mes embarqu\'{e}s" :
[101]10
11COACH will address new embedded systems architectures by allowing the design of  Multi-Core Systems-on-Chip (possibly heterogeneous) on FPGA according to the design constraints and objectives (real-time, low-power). It will permit to design  complex SoC  based on IP cores ((memory, peripherals, network controllers, communication processors), running Embedded Software, as well as an Operating System with associated middleware and API and using hardware accelerator automatically generated. It will also permit to use efficiently different dynamic system management techniques and re-configuration mechanisms.
12
[102]13Axis 2 "Infrastructures pour l'Internet, le calcul intensif ou les services" :
[101]14
15COACH will address High-Performance Computing (HPC) by helping designer to accelerate an application running on a PC by migrating critical parts into a SoC implemented on an FPGA plugged to the PC bus (through a communication link like PCI/X). COACH will reduce the designer effort through the development of tools that translate high level language programs to FPGA configurations. Moreover, Dynamic Partial Reconfiguration will be used for improving HPC performance as well as reducing the required area.
16
[102]17Axis 3 "Robotique et controle/commande" :
[101]18--------------------------------------------------------------------------
19
20
21
[97]22The COACH proposal addresses directly the \emph{Embedded Systems} item of
[67]23the ARPEGE program. It aims to propose solutions to the societal/economical challenges by
[97]24providing SMEs novel design capabilities enabling them to increase their
[25]25design productivity with design exploration and synthesis methods that are placed on top
[97]26of the state-of-the-art methods.
27This project proposes an open-source framework for mapping multi-tasks software applications
28on Field Programmable Gate Array circuits (FPGA).
[99]29%%%
30\parlf
[97]31COACH will contribute to build an open development and run-time
32environment, including communication middleware and tools to support
[25]33developers in the production of embedded software, through all phases of the software lifecycle,
34from requirements analysis until deployment and maintenance.
35More specifically, COACH focuses on:
36\begin{itemize}
37\item High level methods and concepts (esp. requirements and architectural level) for system
38design, development and integration, addressing complexity aspects and modularity.
39\item Open and modular development environments, enabling flexibility and extensibility by
40means of new or sector-specific tools and ensuring consistency and traceability along the
41development lifecycle.
42\item Light/agile methodologies and adaptive workflow providing a dynamic and adaptive
43environment, suitable for co-operative and distributed development.
44\end{itemize}
[99]45%%%
46\parlf
[25]47COACH outcome will contribute to strengthen Europe's competitive position by developing
48technologies and methodologies for product development, focusing (in compliance with the
49scope of the above program) on technologies, engineering methodologies, novel tools,
50methods which facilitate resource use efficiency. The approaches and tools to be developed
51in COACH will enable new and emerging information technologies for the development,
52manufacturing and integration of devices and related software into end-products.
[99]53%%%
54\parlf
[97]55The COACH project will benefit from a number of previous projects:
[99]56\begin{description}
57  \item[SOCLIB]
58    The SoCLib ANR platform (2007-2009) is an open infrastructure developped by 10 academic laboratories
59    and 6 industrial companies.
60    It supports system level virtual prototyping of shared memory, multi-processors
61    architectures, and provides tools to map multi-tasks software application on these
62    architectures, for reliable performance evaluation.
63    The core of this platform is a library of SystemC simulation models for
64    general purpose IP cores such as processors, buses, networks, memories, IO controller.
65    The platform provides also embedded operating systems and software/hardware
66    communication middleware.
67    The synthesisable VHDL models of IPs are not part of the SoCLib platform, and
68    this project enhances SoCLib by providing the synthesisable VHDL models required
69    for FPGA synthesis.
[120]70  \item[ROMA] The ROMA ANR project (http://roma.irisa.fr, 2007-2010)
71    involving IRISA (CAIRN team), LIRMM, CEA List THOMSON France R\&D,
72    proposes to develop a reconfigurable processor, exhibiting high
73    silicon density and power efficiency, able to adapt its computing
74    structure to computation patterns that can be speed-up and/or
75    power efficient.  The ROMA project study a pipeline-based of
76    evolved low-power coarse grain reconfigurable operators to avoid
77    traditional overhead, in reconfigurable devices, related to the
78    interconnection network.  The project will borrow from the ROMA
79    ANR project and the ongoing joint INRIA-STMicro
80    Nano2012 project to adapt existing pattern extraction algorithms
81    and datapath merging techniques to the synthesis of customized
[99]82    ASIP processors.
83  \item[TSAR]
84    The TSAR MEDEA+ project (2008-2010) involving BULL, THALES and the \upmc targets the design of a
85    scalable, coherent shared memory, multi-cores processor architecture, and uses the SoCLib
86    plaform for virtual prototyping. The COACH project will benefit from the synthesizable VHDL
87    models developped in the framework of TSAR (MIPS32 processor core, and RING interconnect).
88  \item[BioWic]
89    On the HPC application side, we also hope to benefit from the experience in
90    hardware acceleration of bioinformatic algorithms/workfows gathered by the
91    CAIRN group in the context of the ANR BioWic project (2009-2011), so as to
92    be able to validate the framework on real-life HPC applications.
93\end{description}
94%%%
95\parlf
[97]96The laboratories involved in the COACH project have a well estabished expertise
97in the following domains:
98\begin{itemize}
[99]99  \item 
100    In the field of High Level Synthesis (HLS), the project
101    leverages on know-how acquired over the last 15 years with the GAUT~\cite{gaut08} project
102    developped by the \ubs laboratory, and with the UGH~\cite{ugh08} project developped
103    by the \upmc and \tima laboratories.
104  \item
105    Regarding system level architecture, the project is based on the know-how
106    acquired by the \upmc and \tima laboratories in the framework of various projects 
[135]107    in the field of communication architectures for shared memory multi-processors systems
108    (COSY~\cite{cosy}, DISYDENT~\cite{disydent05} or DSPIN~\cite{dspin08} of MEDEA-MESA).
109    As an example, the DSPIN project is now used in the TSAR project.
[99]110  \item
111    Regarding Application Specific Instruction Processor (ASIP) design, the
[120]112    CAIRN group at INRIA Rennes -- Bretagne Atlantique benefits from several years of
[99]113    expertise in the domain of retargetable compiler
114    (Armor/Calife~\cite{CODES99} since 1996, and the Gecos
115    compilers~\cite{ASAP05} since 2002).
[97]116\item
[99]117    In the field of compilers, the Compsys group was founded in 2002
118    by several senior researchers with experience in
119    high performance computing and automatic parallelization. They have been
120    among the initiators of the polyhedral model, a theory which serve to
121    unify many parallelism detection and exploitation techniques for regular
122    programs. It is expected that the techniques developped by Compsys for
123    parallelism detection, scheduling, process construction and memory management
124    will be very useful as a front-end for the a high-level synthesis tools.
[97]125\end{itemize}
[99]126%%%
127\parlf
[97]128Finally, it is worth to note that this project cover priorities defined by the commission
[19]129experts in the field of Information Technolgies Society (IST) for Embedded
[97]130Systems: $<<$Concepts, methods and tools for designing systems dealing with systems complexity
[19]131and allowing to apply efficiently applications and various products on embedded platforms,
132considering resources constraints (delais, power, memory, etc.), security and quality
133services$>>$.
Note: See TracBrowser for help on using the repository browser.