Sciweavers

HPCA
2008
IEEE

Prediction of CPU idle-busy activity pattern

13 years 11 months ago
Prediction of CPU idle-busy activity pattern
Real-world workloads rarely saturate multi-core processor. CPU C-states can be used to reduce power consumption during processor idle time. The key unsolved problem is: when and how to use which Cstate. We propose a machine learning prediction method and usage model. We evaluate this model with idle traces collected on dual-core and quad-core processor, and find this method can well predict CPU’s activity pattern at the error level not exceeding 4%. Compared with existing OS C-state policy, it results in 12% additional CPU power saving and 2% performance improvement. In industry, 12% power saving for any processor is very significant improvement. SPECWeb (which we used consists of 3 different benchmarks -- We consistently see doubledigit power saving) is representative “front-end” server workload – it takes >60% DP server market segment share.
Qian Diao, Justin J. Song
Added 29 May 2010
Updated 29 May 2010
Type Conference
Year 2008
Where HPCA
Authors Qian Diao, Justin J. Song
Comments (0)