Zero-Value Caches: Cancelling Loads that Return Zero
Authors:
- Mafijul Md. IslamMIMafijul Md. Islam
- Per StenstromPSPer Stenstrom
- 1 September 2009
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 237-245
Abstract
The speed gap between processor and memory continues to limit performance. To address this problem, we explore the potential of eliminating zero loads - loads accessing memory locations that contain the value ldquozerordquo - to improve performance and energy dissipation. Our study shows that such loads comprise as many as 18% of the total number of dynamic loads. We show that a significant fraction of zero loads ends up on the critical memory-access path in out-of-order cores. We propose a non-speculative microarchitectural technique - zero-value cache (ZVC) - to capitalize on zero loads and explore critical design options of such caches. We show that with modest investment, we can obtain speedups up to 78% and reduce the overall energy dissipation by up to 39%. Most importantly, zero-value caches never cause performance loss.Keywords
This publication has 19 references indexed in Scilit:
- Zero loads2008
- IBM POWER6 microarchitectureIBM Journal of Research and Development, 2007
- NoSQ: Store-Load Communication without a Store QueueIEEE Micro, 2007
- Niagara: A 32-Way Multithreaded Sparc ProcessorIEEE Micro, 2005
- Checkpointed early load retirement2005
- Just say no: benefits of early cache miss determination2003
- The filter cache: an energy efficient memory structure2002
- Automatically characterizing large scale program behavior2002
- Silent stores and store value localityIEEE Transactions on Computers, 2001
- Value locality and load value prediction1996