Utilization of empirically determined energy-optimal CPU-frequencies in a numerical simulation code

Björn Dick*, Andreas Vogel, Dmitry Khabi, Martin Rupp, Uwe Küster, Gabriel Wittum

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

In order to enable exascale computing, concepts for substantial energy savings are required. Dynamic voltage and frequency scaling (DVFS) is widely known to provide suitable energy saving potentials. However, the customarily utilized DVFS mechanism of the Linux kernel determines clock frequencies solely based on an idle time analysis. In contrast to this, we use an empirical approach based on preparatory measurements of the energy consumption at all available frequencies. From the resulting data we deduce energy-optimal frequencies, which are used in subsequent production runs. The described methodology can be deployed with routine granularity to account for varying code characteristics. For evaluation purposes, the approach is applied to the UG4 numerical simulation software. First results exhibit an average energy saving potential of approximately 10 % while increasing the runtime by about 19 %.

Original languageEnglish (US)
Pages (from-to)89-97
Number of pages9
JournalComputing and Visualization in Science
Volume17
Issue number2
DOIs
StatePublished - Apr 1 2015
Externally publishedYes

Keywords

  • DVFS
  • Energy
  • empirical

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Modeling and Simulation
  • General Engineering
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics

Fingerprint

Dive into the research topics of 'Utilization of empirically determined energy-optimal CPU-frequencies in a numerical simulation code'. Together they form a unique fingerprint.

Cite this