DocumentCode :
2573149
Title :
Mixed static/dynamic profiling for dictionary based code compression
Author :
Netto, E. ; Azevedo, R. ; Centoducatte, P. ; Araujo, G.
Author_Institution :
CEFET-RN, Brazil
fYear :
2003
fDate :
19-21 Nov. 2003
Firstpage :
159
Lastpage :
163
Abstract :
Many compression techniques have been proposed to accommodate ever increasing software pieces into restricted memory area in embedded systems. Recently, these techniques have been shown to improve other design constraints like energy and performance. This paper proposes a blended dictionary model based on static/dynamic profiling that lead to best trade-offs on compression, performance and energy savings. We also propose a new dictionary based code compression algorithm, independent of the cache organization and processor, to support our experiments. A mix of benchmarks and MiBench suites reveals that compression ratios of 75% can be obtained while decreasing bus accesses to the cache by 31% for the Leon processor. These results approach simultaneously the best solutions of when using pure static or pure dynamic information based dictionaries techniques.
Keywords :
cache storage; data compression; memory architecture; Leon processor; MiBench suites; benchmarks; blended dictionary model; cache bus access; cache organization; dictionary based code compression; dynamic information based dictionaries; dynamic profiling; embedded systems; processor organization; restricted memory area; software pieces; static information based dictionary; static profiling; Compression algorithms; Data compression; Decoding; Dictionaries; Embedded system; Energy efficiency; Engines; Reduced instruction set computing; Robustness; System-on-a-chip;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
System-on-Chip, 2003. Proceedings. International Symposium on
Print_ISBN :
0-7803-8160-2
Type :
conf
DOI :
10.1109/ISSOC.2003.1267745
Filename :
1267745
Link To Document :
بازگشت