Affordable Access

Access to the full text

An analytical methodology to derive power models based on hardware and software metrics

Authors
  • Dolz, Manuel F.1
  • Kunkel, Julian2
  • Chasapis, Konstantinos1
  • Catalán, Sandra3
  • 1 University of Hamburg, Department of Informatics, Hamburg, 22.527, Germany , Hamburg (Germany)
  • 2 German Climate Computing Center, DKRZ GmbH, Hamburg, 20.146, Germany , Hamburg (Germany)
  • 3 Universitat Jaume I, Depto. de Ingeniería y Ciencia de Computadores, Castellón, 12.071, Spain , Castellón (Spain)
Type
Published Article
Journal
Computer Science - Research and Development
Publisher
Springer Berlin Heidelberg
Publication Date
Sep 11, 2015
Volume
31
Issue
4
Pages
165–174
Identifiers
DOI: 10.1007/s00450-015-0298-8
Source
Springer Nature
Keywords
License
Yellow

Abstract

The use of models to predict the power consumption of a system is an appealing alternative to wattmeters since they avoid hardware costs and are easy to deploy. In this paper, we present an analytical methodology to build models with a reduced number of features in order to estimate power consumption at node level. We aim at building simple power models by performing a per-component analysis (CPU, memory, network, I/O) through the execution of four standard benchmarks. While they are executed, information from all the available hardware counters and resource utilization metrics provided by the system is collected. Based on correlations among the recorded metrics and their correlation with the instantaneous power, our methodology allows (i) to identify the significant metrics; and (ii) to assign weights to the selected metrics in order to derive reduced models. The reduction also aims at extracting models that are based on a set of hardware counters and utilization metrics that can be obtained simultaneously and, thus, can be gathered and computed on-line. The utility of our procedure is validated using real-life applications on an Intel Sandy Bridge architecture.

Report this publication

Statistics

Seen <100 times