Computer Science - Research and Development

, Volume 25, Issue 3–4, pp 155–163 | Cite as

Quantifying power consumption variations of HPC systems using SPEC MPI benchmarks

  • Daniel Hackenberg
  • Robert Schöne
  • Daniel Molka
  • Matthias S. Müller
  • Andreas Knüpfer
Special Issue Paper


The power consumption of an HPC system is not only a major concern due to the huge associated operational cost. It also poses high demands on the infrastructure required to operate such a system. The power consumption strongly depends on the executed workload and is influenced by the system hard- and software and its setup. In this paper we analyze the power consumption of a 32-node cluster across a wide range of parallel applications using the SPEC MPI2007 benchmark. By measuring the variations of the power consumed by different hardware nodes and processes of an applications we lay the ground to extrapolate the energy demand of large parallel HPC systems.


Power consumption Power variation Power extrapolation Energy efficiency SPEC MPI 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    MIMD lattice computation (MILC) collaboration home page (2007)
  2. 2.
    Intel turbo boost technology in Intel core microarchitecture (Nehalem) based processors (2008)
  3. 3.
    Andersson U (1998) Parallelization of a 3D FD-TD code for the Maxwell equations using MPI. In: Kågström B et al. (eds) Applied parallel computing, PARA’98. Lecture notes in computer science, vol 1541. Springer, Berlin, pp 12–19 Google Scholar
  4. 4.
    Bailey D, Harris T, Saphir W, van der Wijngaart R, Woo A, Yarrow M (1995) The NAS parallel benchmarks 2.0. Tech Rep NAS-95-020, NASA Ames Research Center, Moffett Field, CA.
  5. 5.
    Feng W, Cameron K (2006) Green500 run rules: the green500 list: power measurement of high-end clusters. Version 0.1.
  6. 6.
    Feng W, Cameron K (2007) The green500 list: encouraging sustainable supercomputing. Computer 40(12):50–55 CrossRefGoogle Scholar
  7. 7.
    Feng X, Ge R, Cameron KW (2005) Power and energy profiling of scientific applications on distributed systems. In: Proceedings of the 19th IEEE international parallel and distributed processing symposium (IPDPS’05) papers. IEEE Computer Society, Washington, p 34 CrossRefGoogle Scholar
  8. 8.
    Gedney SD (1996) An anisotropic perfectly matched layer absorbing media for the truncation of fdtd lattices. IEEE Trans Antennas Propag 44:1630–1639 CrossRefGoogle Scholar
  9. 9.
    Kamil S, Shalf J, Strohmaier E (2008) Power efficiency in high performance computing. In: IPDPS. IEEE, New York, pp 1–8 Google Scholar
  10. 10.
    Lange KD (2009) Identifying shades of green: the specpower benchmarks. Computer 42:95–97. doi: CrossRefGoogle Scholar
  11. 11.
    Meuer H, Strohmaier E, Dongarra J, Simon H. The Top500 project.
  12. 12.
    Müller MS, van Waveren M, Lieberman R, Whitney B, Saito H, Kumaran K, Baron J, Brantley WC, Parrott C, Elken T, Feng H, Ponder C (2010) Spec mpi2007-an application benchmark suite for parallel systems using mpi. Concurr Comput Pract Exper 22(2):191–205 Google Scholar
  13. 13.
    Plimpton S (1995) Fast parallel algorithms for short-range molecular dynamics. J Comput Phys 117:1–19 zbMATHCrossRefGoogle Scholar
  14. 14.
    Poess M, Nambiar RO (2008) Energy cost, the key challenge of today’s data centers: a power consumption analysis of tpc-c results. Proc VLDB Endow 1(2):1229–1240 Google Scholar
  15. 15.
    Stone JE (1998) An efficient library for parallel ray tracing and animation. PhD thesis, University of Missouri, Rolla Google Scholar
  16. 16.
    zeusmp2 homepage (2007)

Copyright information

© Springer-Verlag 2010

Authors and Affiliations

  • Daniel Hackenberg
    • 1
  • Robert Schöne
    • 1
  • Daniel Molka
    • 1
  • Matthias S. Müller
    • 1
  • Andreas Knüpfer
    • 1
  1. 1.Center for Information Services and High Performance Computing (ZIH)Technische Universität DresdenDresdenGermany

Personalised recommendations