Generalized utility metrics for supercomputers

Open Access
Special Issue Paper

Abstract

The problem of ranking the utility of supercomputer systems arises frequently in situations such as procurements and other types of evaluations of architectures. It is also central for any general ranking of supercomputers such as the Top500. Rankings of computer systems have traditionally solely focused on performance aspects. In recent years restrictions due to power and space requirements of large supercomputers have become very noticeable, which has increased the importance of including these factors in generalized rankings. In this paper we present an overview of the current practice for utility metrics and analyze their shortcomings. We then present and discuss in detail a new concept for a parameterized utility metric for supercomputers, which is based on effective performance, available memory size, actual power consumption, and (if desired) the floor space required for supercomputers. This metric is designed and proposed for augmenting the current Top500 ranking.

Keywords

Computer performance  Utility metrics   Power efficiency   High performance computing market analysis 

References

  1. 1.
    Top500 supercomputer sites. http://www.top500.org (2009)
  2. 2.
    Strohmaier E, Dongarra J, Meuer HW, Simon HD (1999) The marketplace of high-performance computing. Parall Comput 25(13–14):1517–1544CrossRefGoogle Scholar
  3. 3.
    Dongarra J, Luszczek P, Petitet A (2003) The LINPACK benchmark: past, present and future. Concurr Comput Pract Exper 15:1–18CrossRefGoogle Scholar
  4. 4.
    The Green500 List. http://www.green500.org/ (2009)
  5. 5.
    HPC Challenge Benchmark. http://icl.cs.utk.edu/hpcc/ (2009)
  6. 6.
    SPEC: Standard performance evaluation corporation. http://www.spec.org (2009)
  7. 7.
    Feng W-C, Cameron K (2007) The Green500 List: Encouraging Sustainable Supercomputing. Computer 40(12):50–55CrossRefGoogle Scholar
  8. 8.
    Makimoto T, Eguchi K, Yoneyama M (2001) The Cooler the Better: New Directions in the Nomadic Age. Computer 34(4):38–42CrossRefGoogle Scholar
  9. 9.
    STREAM: Sustainable memory bandwidth in high performance computers. http://www.cs.virginia.edu/stream (2009)
  10. 10.
    Oliker L, Carter J, Wehner M, Canning A, Ethier S, Mirin A, Parks D, Worley PH, Kitawaki S, Tsuda Y (2005) Leading Computational Methods on Scalar and Vector HEC Platforms. Proc SC 2005:62Google Scholar
  11. 11.
    Oliker L, Canning A, Carter J, Iancu C, Lijewski M, Kamil S, Shalf J, Shan H, Strohmaier E, Ethier S, Goodale T (2007) Scientific Application Performance on Candidate PetaScale Platforms. Proc IPDPS 2007:1–12Google Scholar
  12. 12.
    Luszczek P, Dongarra J, Koester D, Rabenseifner R, Lucas B, Kepner J, McCalpin J, Bailey D, Takahashi D (2005) Introduction the the HPC challenge benchmark suite. Available at http://www.hpcchallenge.org/pubs/
  13. 13.
    Kramer W, Shalf J, Strohmaier E (2005) The NERSC Sustained System Performance (SSP) Metric. Lawrence Berkeley National Laboratory. Paper LBNL-58868. http://repositories.cdlib.org/lbnl/LBNL-58868

Copyright information

© The Author(s) 2009

Authors and Affiliations

  1. 1.Future Technologies GroupLawrence Berkeley National LaboratoryBerkeleyUSA

Personalised recommendations