Changeset 38 for anr/section-2.tex
- Timestamp:
- Jan 19, 2010, 11:01:35 AM (14 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
anr/section-2.tex
r32 r38 6 6 complexity of design problems. Such methods, addressing these challenges starting from high levels of 7 7 abstraction, will have to perform large solution space explorations both for software and (possibly 8 reconfigurable) h rdware, involving almost marginaldesign effort and offering a high predictability of results9 with respect to cost - and performance-objectives.8 reconfigurable) hardware, reducing the design effort and offering a high predictability of results 9 with respect to cost and performance objectives. 10 10 Current design methodologies provide quite low-level abstraction capabilities. However in a few years 11 11 from now tens of programmable processors will be embedded in an IC with more than 100M 12 transistors adding to the complexity of the problem of designing such systems. Taking into account 13 that the complexity of the software part is increasing at an even faster rate, current solutions for 14 design space exploration, mainly manually based, by no means do supply an adequate performance. 12 transistors, therefore adding to the complexity of the problem of designing such systems. 13 Taking into account that the complexity of the software part is increasing at an even 14 faster rate, current solutions for design space exploration, mainly manually based, by no 15 means do supply an adequate efficiency. 15 16 Consequently, there is an urgent need to leverage system level 16 17 exploration through the use of a high level specification of the application and an early design … … 22 23 \\ 23 24 Thus, new system-level design flows need to be developed, enabling the exploration of an application 24 independently of the implementation, almost at the beginning of the design process. A25 fundamental element of this evolution is the definition of abstraction layers that should allow the26 systematic re-use of software and hardware components at the system level driven by performance estimation 27 and analysis. In this context, COACH will combine modeling and estimation methods and 28 compilers and design space exploration techniques. This approach will cause a real breakthrough in 29 the embedded system design methodology, i.e. one of the radical innovations.25 independently of the implementation, almost at the beginning of the design process. 26 A fundamental element of this evolution is the definition of abstraction layers that should allow the 27 performance driven re-use of software and hardware components at the system level. 28 In this context, COACH will combine modeling and estimation methods and compilers and 29 design space exploration techniques. This approach will be a radical innovation in 30 embedded system design methodology. 30 31 \\ 31 The reason is that COACH precedes the use ofhigh-level design tools in the embedded32 systems design flow s. In that way, it will make possible a real and efficiently combined33 exploitation of high-level synthesis tools, paralleli sing approaches and compilers, already32 The reason is that the COACH framework is applied before high-level design tools in the embedded 33 systems design flow. In that way, it will make possible a real and efficiently combined 34 exploitation of high-level synthesis tools, parallelizing approaches and compilers, already 34 35 available on the market. These tools and approaches are not yet massively adopted, precisely 35 because this decisivedesign step is missing. COACH will indeed permit (i) to predict and36 because this preliminary design step is missing. COACH will indeed permit (i) to predict and 36 37 control implementation optimizations, (ii) to target multiple implementation technologies 37 38 (and thus the associated tools) from a unique specification and (iii) to efficiently integrate high … … 43 44 a lower global cost. 44 45 \par 45 To get an efficient embedded system, designer has to take into account46 application characteristics when it chooses one of the formertechnologies.47 This choice is not easy and in most cases designer has to try different46 To get an efficient embedded system, the system designer has to take into account 47 application characteristics when it chooses one of the available technologies. 48 This choice is not easy and in most cases the designer has to try different 48 49 technologies to retain the most adapted one. 49 50 \\ 50 51 The first objective of COACH is to provide an open-source framework to 51 design embedded system on FPGA device .52 COACH framework allowsdesigner to explore various software/hardware52 design embedded system on FPGA devices. 53 The COACH framework allows the designer to explore various software/hardware 53 54 partitions of the target application, to run timing and functional 54 55 simulations and to generate automatically both the software and the … … 65 66 Micro-architectural exploration: When hardware components are required, the 66 67 HLS tools of the framework generate them automatically. At this stage the 67 framework provides various HLS tools allowingthe micro-architectural space68 framework provides various HLS tools that allow the micro-architectural space 68 69 design exploration. The exploration criteria also are throughput, latency 69 70 and power consumption. … … 74 75 75 76 \item 76 Performance measurement: For each point of design space exploration,77 metrics of criteriaare available such as throughput, latency, power77 Performance measurement: For each point in the design space, 78 figures of merit are available such as throughput, latency, power 78 79 consumption, area, memory allocation and data locality. They are evaluated 79 using virtual prototyping, estimation or analy sing methodologies.80 using virtual prototyping, estimation or analyzing methodologies. 80 81 \item 81 Targeted hardware technology: The COACH description of system is82 independent of the FPGA family. Every point of the design exploration82 Targeted hardware technology: The COACH description of a system is 83 independent of the FPGA family. Every point of the design 83 84 space can be implemented on any FPGA having the required resources. 84 85 Basically, COACH handles both Altera and Xilinx FPGA families. … … 86 87 As an extension of embedded system design, COACH deals also with High 87 88 Performance Computing (HPC). 88 In HPC, the kind of targeted application is an existing one running on PC.89 COACHhelps designer to accelerate it by migrating critical parts into a90 SoC implemented on a FPGA plugged to the PC bus.89 In HPC, the kind of targeted application is an existing one running on a PC. 90 The COACH framework helps designer to accelerate it by migrating critical parts into a 91 SoC implemented on an FPGA plugged to the PC bus. 91 92 \par 92 COACH is the result of the will of several laborator y to unify their know93 howand skills in the following domains: Operating system and hardware94 communication ( TIMA, SITI), SoC and MPSoC (LIP6 and TIMA), ASIP (IRISA) and95 HLS ( LIP6, Lab-STIC and LIP).93 COACH is the result of the will of several laboratories to unify their knowhow 94 and skills in the following domains: Operating system and hardware 95 communication (\tima, \citi), SoC and MPSoC (\upmc and \tima), ASIP (\irisa) and 96 HLS (\upmc, \ubs) and compilation (\irisa, \lip). 96 97 The project objective is to integrate these various domains into a unique 97 98 free framework (licence ...) masking as much as possible these domains and
Note: See TracChangeset
for help on using the changeset viewer.