Changeset 356 for anr/section-etat-de-art.tex
- Timestamp:
- Feb 6, 2011, 2:29:09 PM (14 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
anr/section-etat-de-art.tex
r339 r356 25 25 By adapting architecture to the software, % (the opposite is done in the others families) 26 26 FPGAs architectures enable better performance 27 (typically between x10 and x100 accelerations)27 (typically an acceleration factor between 10 and 100) 28 28 while using smaller size and less energy (and heat). 29 29 However, using FPGAs presents significant challenges~\cite{hpc06a}. … … 128 128 \item The parallelism is extracted from the 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 are techniques such as poly edric130 must re-write the algorithmic specification while there are techniques such as polyhedral 131 131 transformations to increase the intrinsic parallelism, 132 132 \item While they support limited loop transformations like loop unrolling and loop … … 213 213 Since hardware is inherently parallel, finding parallelism in sequential 214 214 programs in an important prerequisite for HLS. The large FPGA chips of 215 today can accom odate much more parallelism than is available in basic blocks.215 today can accommodate much more parallelism than is available in basic blocks. 216 216 The polyhedral model is the ideal tool for finding more parallelism in 217 217 loops. … … 219 219 As a side effect, it has been observed that the polyhedral model is a useful 220 220 tool for many other optimization, like memory reduction and locality 221 improvement. It should be noted221 improvement. It should be noted 222 222 that the polyhedral model \emph{stricto sensu} applies only to 223 223 very regular programs. Its extension to more general programs is … … 251 251 further releases. 252 252 \parlf 253 In IP-XACT the flow automation and data cons tistency is ensured by generators, which253 In IP-XACT the flow automation and data consistency is ensured by generators, which 254 254 are program modules that process IP-XACT XML data into something useful 255 255 for the design. They are key portable mechanism for encapsulating specialist design 256 256 knowledge and enable designers to deploy specialist knowledge in their design. It is 257 257 always possible to create generators in order to link several design or analysis tools 258 around a centric representation of meta data in IP-XACT. This kind of XML schema for259 meta data management is a good solution for the federation of heterogeneous design domains258 around a centric representation of meta-data in IP-XACT. This kind of XML schema for 259 meta-data management is a good solution for the federation of heterogeneous design domains 260 260 (models, tools, languages, methodologies, etc.). 261 261 262 %\subsubsection{High Performance Computing}263 %Accelerating high-performance computing (HPC) applications with field-programmable264 %gate arrays (FPGAs) can potentially improve performance.265 %However, using FPGAs presents significant challenges~\cite{hpc06a}.266 %First, the operating frequency of an FPGA is low compared to a high-end microprocessor.267 %Second, based on Amdahl law, HPC/FPGA application performance is unusually sensitive268 %to the implementation quality~\cite{hpc06b}.269 %Finally, High-performance computing programmers are a highly sophisticated but scarce270 %resource. Such programmers are expected to readily use new technology but lack the time271 %to learn a completely new skill such as logic design~\cite{hpc07a} .272 %\\273 %HPC/FPGA hardware is only now emerging and in early commercial stages,274 %but these techniques have not yet caught up.275 %Thus, much effort is required to develop design tools that translate high level276 %language programs to FPGA configurations.277
Note: See TracChangeset
for help on using the changeset viewer.