Title :
Compiler-directed leakage energy reduction for instruction scratch-pad memories
Author :
Yijie Huangfu ; Wei Zhang
Author_Institution :
Dept. of Electr. & Comput. Eng., Virginia Commonwealth Univ., Richmond, VA, USA
Abstract :
Scratch-Pad Memories (SPMs) have been increasingly used in embedded systems, for which it is important to reduce the leakage energy consumption. While there are several techniques to reduce the leakage energy for cache memories, those techniques may result in significant performance overheads for instruction SPMs. In this paper, we study a compiler-based approach to reducing the instruction SPM leakage energy efficiently, which can also minimize the performance overhead. Our evaluation indicates that the compiler-based approach is superior to periodical or bank-based methods, all of which use drowsy techniques. On average, the compiler-based method can reduce the SPM leakage energy by nearly 89.82%, with only 0.25% performance overhead.
Keywords :
SRAM chips; power aware computing; program compilers; bank-based methods; cache memories; compiler-based approach; compiler-directed leakage energy consumption reduction; drowsy techniques; embedded systems; instruction SPM leakage energy; instruction scratch-pad memories; Benchmark testing; Energy consumption; Equations; Mathematical model; Memory management; Program processors; Resource management; Compiler; Drowsy Caches; Leakage Energy Consumption; Performance; Scratch-Pad Memories (SPMs);
Conference_Titel :
Quality Electronic Design (ISQED), 2014 15th International Symposium on
Conference_Location :
Santa Clara, CA
Print_ISBN :
978-1-4799-3945-9
DOI :
10.1109/ISQED.2014.6783352