Learning based DVFS for simultaneous temperature, performance and energy management

Hao Shen, Jun Lu, Qinru Qiu

Research output: Chapter in Book/Entry/PoemConference contribution

59 Scopus citations

Abstract

Dynamic voltage and frequency scaling (DVFS) has been widely used for energy reduction in the modern processors. How to select the optimal frequency that minimizes energy dissipation for the given performance constraint at runtime is a nontrivial problem. The problem becomes more complicated if temperature needs to be constrained (or minimized) simultaneously. The temperature, performance and energy have different nonlinear relationships with frequency/voltage scaling ratio and this relationship is closely related to the characteristics of hardware and applications. In this paper, we design a reinforcement learning algorithm to tackle the problem of simultaneous temperature, performance and energy management. The proposed approach allows continuous tradeoff among these three quality measurement of a computer system. It also enables us to set two of the measurements as constraints and optimize the third one. The proposed approach is validated on an Intel Core 2 processor running Linux system.

Original languageEnglish (US)
Title of host publicationProceedings of the 13th International Symposium on Quality Electronic Design, ISQED 2012
Pages747-754
Number of pages8
DOIs
StatePublished - 2012
Event13th International Symposium on Quality Electronic Design, ISQED 2012 - Santa Clara, CA, United States
Duration: Mar 19 2012Mar 21 2012

Publication series

NameProceedings - International Symposium on Quality Electronic Design, ISQED
ISSN (Print)1948-3287
ISSN (Electronic)1948-3295

Other

Other13th International Symposium on Quality Electronic Design, ISQED 2012
Country/TerritoryUnited States
CitySanta Clara, CA
Period3/19/123/21/12

Keywords

  • DVFS
  • dynamic voltage and frequency scaling
  • energy
  • enhancement learning
  • performance
  • temperature

ASJC Scopus subject areas

  • Hardware and Architecture
  • Electrical and Electronic Engineering
  • Safety, Risk, Reliability and Quality

Fingerprint

Dive into the research topics of 'Learning based DVFS for simultaneous temperature, performance and energy management'. Together they form a unique fingerprint.

Cite this