DocumentCode
3262795
Title
Lazy BTB: reduce BTB energy consumption using dynamic profiling
Author
Chang, Yen-Jen
Author_Institution
Dept. of Comput. Sci., National Chung-Hsing Univ., Taichung
fYear
2006
fDate
24-27 Jan. 2006
Abstract
In this paper, we propose an alternative BTB design, called lazy BTB, to reduce the BTB energy consumption by filtering out the redundant lookups. The most distinct feature of the lazy BTB is that it dynamically profiles the taken traces during program execution. Unlike the traditional design in which the BTB has to be looked up every instruction fetch, by introducing an additional field to record the trace information, our design can achieve the goal of one BTB lookup per taken trace. The experimental results show that with a negligible performance degradation the lazy BTB can reduce the BTB energy consumption by about 77% on average for the MediaBench applications
Keywords
cache storage; content-addressable storage; dynamic programming; network synthesis; performance evaluation; table lookup; MediaBench application; dynamic profiling; energy consumption; filtering; lazy BTB; performance degradation; redundant lookup; trace information; Associative memory; Decoding; Degradation; Energy consumption; Filtering; Hardware; Hazards; Pipeline processing; Random access memory; Speech synthesis;
fLanguage
English
Publisher
ieee
Conference_Titel
Design Automation, 2006. Asia and South Pacific Conference on
Conference_Location
Yokohama
Print_ISBN
0-7803-9451-8
Type
conf
DOI
10.1109/ASPDAC.2006.1594803
Filename
1594803
Link To Document