MediaBench: a tool for evaluating and synthesizing multimedia and communicatons systems
MICRO 30 Proceedings of the 30th annual ACM/IEEE international symposium on Microarchitecture
Proceedings of the 2002 IEEE/ACM international conference on Computer-aided design
Intraprogram dynamic voltage scaling: Bounding opportunities with analytic modeling
ACM Transactions on Architecture and Code Optimization (TACO)
MiBench: A free, commercially representative embedded benchmark suite
WWC '01 Proceedings of the Workload Characterization, 2001. WWC-4. 2001 IEEE International Workshop
Real-time loop scheduling with energy optimization via DVS and ABB for multi-core embedded system
EUC'07 Proceedings of the 2007 international conference on Embedded and ubiquitous computing
Hi-index | 0.00 |
Traditionally, active power has been the primary source of power dissipation in CMOS designs. Although, leakage power is becoming increasingly more important as technology feature sizes continue to shrink, traditioinal power optimization techniques often neglect its contribution to total system power. In this paper, we present a power-aware compilation methodology that targets an embedded processor with both dynamic voltage scaling (DVS) and adaptive body biasing (ABB) capabilities. Our technique has the unique advantage of optimizing design power by jointly optimizing dynamic and leakage power dissipation. Considering the delay and energy penalty of swithching between processor modes, our compiler generates code with minimum power consumption under deadline constraints. Compared to not performing any optimization, or using DVS alone, our technique improves the power consumption of a number of embedded application kernels by 26%, and 14%, respectively.