Predictive Power Management for Multi-Core Processors - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2010

Predictive Power Management for Multi-Core Processors

Abstract

Predictive power management provides reduced power consumption and increased performance compared to reactive schemes. It effectively reduces the lag between workload phase changes and changes in power adaptations since adaptations can be applied immediately before a program phase change. To this end we present the first analysis of prediction for power management under SYSMark2007. Compared to traditional scientific/computing benchmarks, this workload demonstrates more complex core active and idle behavior. We analyze a table based predictor on a quad-core processor. We present an accurate run-time power model that accounts for fine-grain temperature and voltage variation. By predictively borrowing power from cores, our approach provides an average speedup of 7.3% in SYSMark2007.
Fichier principal
Vignette du fichier
WEED-bircher.pdf (326.41 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

inria-00492839 , version 1 (17-06-2010)

Identifiers

  • HAL Id : inria-00492839 , version 1

Cite

William Lloyd Bircher, Lizy John. Predictive Power Management for Multi-Core Processors. WEED 2010 - Workshop on Energy-Efficient Design, Jun 2010, Saint Malo, France. ⟨inria-00492839⟩

Collections

WEED2010 ISCA2010
83 View
603 Download

Share

Gmail Facebook Twitter LinkedIn More