Power-capped DVFS and thread allocation with ANN models on modern NUMA systems

Satoshi Imamura, Hiroshi Sasaki, Koji Inoue, Dimitrios S. Nikolopoulos

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

Abstract

Power capping is an essential function for efficient power budgeting and cost management on modern server systems. Contemporary server processors operate under power caps by using dynamic voltage and frequency scaling (DVFS). However, these processors are often deployed in non-uniform memory access (NUMA) architectures, where thread allocation between cores may significantly affect performance and power consumption. This paper proposes a method which maximizes performance under power caps on NUMA systems by dynamically optimizing two knobs: DVFS and thread allocation. The method selects the optimal combination of the two knobs with models based on artificial neural network (ANN) that captures the nonlinear effect of thread allocation on performance. We implement the proposed method as a runtime system and evaluate it with twelve multithreaded benchmarks on a real AMD Opteron based NUMA system. The evaluation results show that our method outperforms a naive technique optimizing only DVFS by up to 67.1%, under a power cap.

Original languageEnglish
Title of host publication2014 32nd IEEE International Conference on Computer Design, ICCD 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages324-331
Number of pages8
ISBN (Electronic)9781479964925
DOIs
Publication statusPublished - Dec 3 2014
Event32nd IEEE International Conference on Computer Design, ICCD 2014 - Seoul, Korea, Republic of
Duration: Oct 19 2014Oct 22 2014

Publication series

Name2014 32nd IEEE International Conference on Computer Design, ICCD 2014

Other

Other32nd IEEE International Conference on Computer Design, ICCD 2014
CountryKorea, Republic of
CitySeoul
Period10/19/1410/22/14

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Power-capped DVFS and thread allocation with ANN models on modern NUMA systems'. Together they form a unique fingerprint.

Cite this