Computer Science - Research and Development

, Volume 29, Issue 3, pp 221–229

Automatic detection of power bottlenecks in parallel scientific applications

  • María Barreda
  • Sandra Catalán
  • Manuel F. Dolz
  • Rafael Mayo
  • Enrique S. Quintana-Ortí
Special Issue Paper

DOI: 10.1007/s00450-013-0242-8

Cite this article as:
Barreda, M., Catalán, S., Dolz, M.F. et al. Comput Sci Res Dev (2014) 29: 221. doi:10.1007/s00450-013-0242-8

Abstract

In this paper we present an extension of the pmlib framework for power-performance analysis that permits a rapid and automatic detection of power sinks during the execution of concurrent scientific workloads. The extension is shaped in the form of a multithreaded Python module that offers high reliability and flexibility, rendering an overall inspection process that introduces low overhead. Additionally, we investigate the advantages and drawbacks of the RAPL power model, introduced in the Intel Xeon “Sandy-Bridge” CPU, versus a data acquisition system from National Instruments.

Keywords

Power efficiencyHigh performance computingProfiling and tracing power consumptionScientific applications

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • María Barreda
    • 1
  • Sandra Catalán
    • 1
  • Manuel F. Dolz
    • 2
  • Rafael Mayo
    • 1
  • Enrique S. Quintana-Ortí
    • 1
  1. 1.Depto. de Ingeniería y Ciencia de ComputadoresUniv. Jaume ICastellónSpain
  2. 2.Department of InformaticsUniversity of HamburgHamburgGermany