The filter cache: an energy efficient memory structure
MICRO 30 Proceedings of the 30th annual ACM/IEEE international symposium on Microarchitecture
Power considerations in the design of the Alpha 21264 microprocessor
DAC '98 Proceedings of the 35th annual Design Automation Conference
ISLPED '98 Proceedings of the 1998 international symposium on Low power electronics and design
ISLPED '99 Proceedings of the 1999 international symposium on Low power electronics and design
Gated-Vdd: a circuit technique to reduce leakage in deep-submicron cache memories
ISLPED '00 Proceedings of the 2000 international symposium on Low power electronics and design
Cache decay: exploiting generational behavior to reduce cache leakage power
ISCA '01 Proceedings of the 28th annual international symposium on Computer architecture
Hardware and Software Techniques for Controlling DRAM Power Modes
IEEE Transactions on Computers
The Future of Systems Research
Computer
Power: A First Class Design Constraint for Future Architecture and Automation
HiPC '00 Proceedings of the 7th International Conference on High Performance Computing
Cache-Line Decay: A Mechanism to Reduce Cache Leakage Power
PACS '00 Proceedings of the First International Workshop on Power-Aware Computer Systems-Revised Papers
HPCA '01 Proceedings of the 7th International Symposium on High-Performance Computer Architecture
Power reduction techniques for microprocessor systems
ACM Computing Surveys (CSUR)
Hi-index | 0.00 |
Both power and performance are important design parameters of the present day processors. This paper explores an integrated software and circuit level technique to reduce leakage power in L1 instruction caches of high performance microprocessors, by eliminating basic blocks from the cache, as soon as they are dead. The effect of this dead block elimination in cache on both the power consumption of the I-cache and the performance of the processor is studied. Identification of basic blocks is done by the compiler from the control flow graph of the program. This information is conveyed to the processor, by annotating the first instruction of selected basic blocks. During execution, the blocks that are not needed further are traced and invalidated and the lines occupied by them are turned off. This mechanism yields an average of about 5% to 16% reduction, in the energy consumed for different sizes of I-cache, for a set of the SPEC CPU 2000 benchmarks [16], without any performance degradation.