- Timestamp:
- Jan 20, 2011, 12:50:48 PM (14 years ago)
- Location:
- anr
- Files:
-
- 8 edited
Legend:
- Unmodified
- Added
- Removed
-
anr/annexe-cv.tex
r318 r319 84 84 \end{itemize} 85 85 \end{cvenv} 86 % 87 \begin{cvenv} 88 {{Alias}{Christophe}{33}} 89 {Junior Researcher, INRIA} 90 {PhD Thesis, 2005} 91 {{%number of pubs 92 }{%citations 93 }} 94 \item[Course of Lectures]\mbox{} 95 Compilation, Architecture. 96 \item[Projects]\mbox{} 97 \begin{itemize} 98 \item S2S4HLS, 2009-- 99 \end{itemize} 100 \item[Other Work Experience]\mbox{} 101 \begin{itemize} 102 \item Post docs at ENS Lyon/LIP and Ohio State U ? 103 \end{itemize} 104 \end{cvenv} -
anr/section-1.tex
r316 r319 31 31 such as ATOM E600C (Intel). 32 32 Probably in few years, one can expect that such chips will become current and even standard 33 general purpose CPU cores will contains a configurable area making explode the low and medium volume 34 markets of digital systems. 33 general purpose CPU cores will contains a configurable area% 34 %paul 35 bringing an explosion in low and medium volume markets. 35 36 \parlf 36 37 The objective of COACH is to provide an integrated design flow for the design of -
anr/section-2.tex
r311 r319 72 72 unadapted to the current GPU based solutions. 73 73 \parlf 74 Coach generates SoC which is part of larger system. Thus it 's important to take in account the existing industrial design flow. For this reason COACH will use the IP-XACT IEEE 1685 standard for packaging these generated SoC.74 Coach generates SoC which is part of larger system. Thus it is important to take in account the existing industrial design flow. For this reason COACH will use the IP-XACT IEEE 1685 standard for packaging these generated SoC. 75 75 \begin{center}\begin{minipage}{.8\linewidth}\textit{ 76 76 The third objective of COACH is to facilitate the integration of generated SoC in global system design flow. -
anr/section-consortium-people.tex
r315 r319 44 44 45 45 \peopletabularentry{\lip} 46 responsible & Feautrier & Paul & professor & ... & ... & ...\\\hline47 ... & ... & ... & ... & ... & ... & ... \\\hline 48 ... & ... & ... & ... & ... & ... & ... \\\hline46 responsible & Alias & Christophe & junior reasearcher & HLS & ... & Memory management, HLS \\\hline 47 contributor & Feautrier & Paul & emeritus professor & Compil. & ... & Process scheduling and building. 48 Automatic Parallelization \\ \hline 49 49 50 50 \peopletabularentry{\tima} -
anr/section-etat-de-art.tex
r315 r319 78 78 \\ 79 79 In the opposite, SOPC Builder~\cite{spoc-builder} from \altera and \xilinx 80 Platform Studio XPS from \xilinx allows to describe a system, to synthesi sit,80 Platform Studio XPS from \xilinx allows to describe a system, to synthesize it, 81 81 to program it into a target FPGA and to upload a software application. 82 82 Both SOPC Builder and XPS, allow designers to select and parameterize components from … … 128 128 \item The parallelism is extracted from initial specification. 129 129 To get more parallelism or to reduce the amount of required memory in the SoC, the user 130 must re-write the algorithmic specification while there istechniques such as polyedric130 must re-write the algorithmic specification while there are techniques such as polyedric 131 131 transformations to increase the intrinsic parallelism, 132 132 \item While they support limited loop transformations like loop unrolling and loop 133 pipelining, current HLS tools do not provide support for design space exploration neither134 through automatic loop transformations nor through memory mapping,133 pipelining, current HLS tools do not provide support for design space exploration, either 134 through automatic loop transformations or through memory mapping, 135 135 \item Despite having the same input language (C/C++), they are sensitive to the style in 136 which the algorithm dis written. Consequently, engineering work is required to swap from136 which the algorithm is written. Consequently, engineering work is required to swap from 137 137 a tool to another, 138 138 \item They do not respect accurately the frequency constraint when they target an FPGA device. … … 146 146 \label{soa:asip} 147 147 ASIP (Application-Specific Instruction-Set Processor) are programmable 148 processors in which both the instruction and the micro architecture have148 processors in which both the instruction set and the micro architecture have 149 149 been tailored to a given application domain or to a 150 150 specific application. This specialization usually offers a good compromise … … 220 220 tool for many other optimization, like memory reduction and locality 221 221 improvement. Another point is 222 that the polyhedral domain\emph{stricto sensu} applies only to222 that the polyhedral model \emph{stricto sensu} applies only to 223 223 very regular programs. Its extension to more general programs is 224 224 an active research subject. … … 245 245 are members of the consortium and the board is incorporating top actors 246 246 (STM, NXP, TI, ARM, FREESCALE, LSI, Mentor, Synopsys and Cadence), ensuring the 247 wide adoption by industry. Initiatives have already work for extending this standard 247 wide adoption by industry. Initiatives have already% work for (paul) 248 attempted to extend this standard 248 249 to AMS IPs packaging domain (MEDEA+ Beyond Dreams Project) and to Hardware Dependent 249 250 Software layers (MEDEA+ SoftSoc project) and Accellera is reusing these results for -
anr/section-issues.tex
r312 r319 33 33 thanks to the developpment of design methodologies and tools for embedded systems. 34 34 Unfortunately, the Non Recurring Engineering (NRE) costs involded in the design 35 and manufacturing ASICs is very high.35 and manufacturing of ASICs is very high. 36 36 An IC foundry costs several billions of euros and the fabrication of a specific circuit 37 37 costs several millions. For example a conservative estimate for a 65nm ASIC project is 10 38 million USD. Consequently, it is more and more unaffordable to design and fabricate ASICs for low and medium38 million USD. Consequently, it is more and more unaffordable to design and fabricate ASICs for low and medium 39 39 volume markets and the new trend for building the new generation products will be multi processors SoCs and programmable logic for co-processsing. 40 40 \\ 41 41 According to a market survey (J-M. Chery, CTO ST Microelectronics at European NanoelectronicsForum 2010), the global growth is 30 Billons\$ between 2009-2013 for multimedia and communication sectors; this is 6 times more than all other domains like security, home automation, health. 42 The predominance of market of multimedia and communication sectors results in the fact that they are mainly mass market. 42 The predominance%paul 43 the multimedia and communication sectors 44 %results paul 45 are due to their being predominently a mass market. 43 46 % 44 47 \subsubsection*{FPGAs and Embedded Systems} … … 84 87 \subsubsection*{COACH's contribution to this evolution} 85 88 Nowadays, there are no commercial or academic tools covering the whole design flow 86 from the system level specification to the bitstream generation neither for embedded system design87 nor for HPC.89 from the system level specification to the bitstream generation, either for embedded system design 90 or for HPC. 88 91 \begin{center}\begin{minipage}{.9\linewidth}\textit{ 89 92 The aim of the COACH project is to integrate all these design steps into a single design framework … … 106 109 to launch start-ups in software engineering. 107 110 \\ 108 So this may increase the total amount of engineers working in this domain: today in France the total is only 26,000 in which 16,000 in big companies \cite{rapport-ministere}. 111 So this may increase the total%amount (paul) 112 number of engineers working in this domain: today in France the total is only 26,000 of which 16,000 are in big companies \cite{rapport-ministere}. -
anr/section-objectif.tex
r313 r319 12 12 The design steps are presented figure~\ref{coach-flow}. 13 13 The end-user input is 14 either a HPC application (an application running on a PC that must be accelerate ),14 either a HPC application (an application running on a PC that must be accelerated), 15 15 or an embedded application (a standalone application), 16 16 or a sub-system application of a larger design. … … 74 74 and the HLS tools of COACH will support a common language and coding style 75 75 to avoid re-engineering by the designer. 76 COACH will provide a tool which will automatically explore the micro-architectural77 design space of coprocessor.76 COACH will provide a tool which will automatically explore the%paul 77 coprocessor micro-architectural design space. 78 78 \item[\textit{High-level code transformation}]: 79 79 COACH will allow to optimize the memory usage, to enhance the parallelism through -
anr/section-position.tex
r315 r319 27 27 environment, including communication middleware and tools to support 28 28 developers in the production of embedded software, through all phases of the software lifecycle, 29 from requirements analysis down to deployment and maintenance.29 from requirements analysis down to deployment and maintenance. 30 30 More specifically, COACH focuses on: 31 31 \begin{itemize} … … 146 146 to it since 2003 and Magillem tool suite is used for its validation. Magillem is used in 147 147 industrial production flows of ST, NXP, TI, Qualcomm, and system integrators like Thales, 148 Astrium, Thomson, etc. what guarantees a strong alignement on customers needs and enhanceresults exploitation.148 Astrium, Thomson, etc. This guarantees a strong alignement on customers needs and enhanced results exploitation. 149 149 150 150 \end{itemize} … … 152 152 \subsubsection*{Relevance to the call axis} 153 153 This project answer to the global statement of the call "INGENIERIE NUMERIQUE ET SECURITE (INS)" by proposing 154 methods and tools for the design of application to be r an on platforms of the next generation. Results will be gained in term of productivity,154 methods and tools for the design of application to be run on platforms of the next generation. Results will be gained in term of productivity, 155 155 time-to-market (automation and code generation) and safety (management of high level sepcifications down to implementation). 156 156 In this call, the COACH project totally fulfills the objectives of the axis 2 "METHODES, … … 161 161 based on IP cores (memory, peripherals...), 162 162 running Embedded Software, as well as an Operating System with associated middleware and 163 API and using hardware accelerator automatically generated. It will also permit to use163 API and using automatically generated hardware accelerators. It will also permit to use 164 164 efficiently different dynamic system management techniques and re-configuration mechanisms. 165 165 The results will be tailored in order to be integrated in standard design flow of critical systems.
Note: See TracChangeset
for help on using the changeset viewer.