Exploiting dynamic phase distance mapping for phase-based tuning of embedded systems

Tosiron Adegbija, Ann Gordon-Ross

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Phase-based tuning increases optimization potential by configuring system parameters for application execution phases. Previous work proposed phase distance mapping (PDM), which relied on extensive a priori analysis of executing applications to dynamically estimate the best configuration using the correlation between phases. We propose DynaPDM, a new dynamic phase distance mapping methodology that eliminates a priori designer effort, dynamically analyzes phases, and determines the best configurations, yielding average energy delay product savings of 28% - an 8% improvement on PDM - and configurations within 1% of the optimal.

Original languageEnglish (US)
Title of host publication2013 IEEE 31st International Conference on Computer Design, ICCD 2013
PublisherIEEE Computer Society
Pages363-368
Number of pages6
ISBN (Print)9781479929870
DOIs
StatePublished - 2013
Externally publishedYes
Event2013 IEEE 31st International Conference on Computer Design, ICCD 2013 - Asheville, NC, United States
Duration: Oct 6 2013Oct 9 2013

Publication series

Name2013 IEEE 31st International Conference on Computer Design, ICCD 2013

Conference

Conference2013 IEEE 31st International Conference on Computer Design, ICCD 2013
Country/TerritoryUnited States
CityAsheville, NC
Period10/6/1310/9/13

Keywords

  • Cache tuning
  • configurable hardware
  • dynamic reconfiguration
  • energy delay product savings
  • phase-based tuning

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Hardware and Architecture

Fingerprint

Dive into the research topics of 'Exploiting dynamic phase distance mapping for phase-based tuning of embedded systems'. Together they form a unique fingerprint.

Cite this