Advertisement

Reproducible Measurements of MPI Performance Characteristics

  • William Gropp
  • Ewing Lusk
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1697)

Abstract

In this paper we describe the difficulties inherent in making accurate, reproducible measurements of message-passing performance. We describe some of the mistakes often made in attempting such measurements and the consequences of such mistakes. We describe mpptest, a suite of performance measurement programs developed at Argonne National Laboratory, that attempts to avoid such mistakes and obtain reproducible measures of MPI performance that can be useful to both MPI implementors and MPI application writers. We include a number of illustrative examples of its use.

Keywords

Message Size Reproducible Measurement Message Length Easure Time Ping Pong 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Benchweb. World Wide Web. http://www.netlib.org/benchweb/.
  2. 2.
    Parkbench Committee. Public international benchmarks for parallel computers. Scientific Programming, 3(2):101–146, 1994. Report 1.Google Scholar
  3. 3.
    W. Gropp and E. Lusk. A high-performance MPI implementation on a shared-memory vector supercomputer. Parallel Computing, 22(11):1513–1526, January 1997.zbMATHCrossRefGoogle Scholar
  4. 4.
    W. D. Gropp and E. Lusk. Experiences with the IBM SP1. IBM Systems Journal, 34(2):249–262, 1995.CrossRefGoogle Scholar
  5. 5.
    J. Piernas, A. Flores, and J. M. García. Analyzing the performance of MPI in a cluster of workstations based on fast Ethernet. In Marian Bubak, Jack Dongarra, and Jerzy Waśniewski, editors, Recent advances in Parallel Virtual Machine and Message Passing Interface, volume 1332 of Lecture Notes in Computer Science, pages 17–24. Springer, 1997. 4th European PVM/MPI Users’ Group Meeting.Google Scholar
  6. 6.
    Michael Resch, Holger Berger, and Thomas Boenisch. A comparision of MPI performance on different MPPs. In Marian Bubak, Jack Dongarra, and Jerzy Waśniewski, editors, Recent advances in Parallel Virtual Machine and Message Passing Interface, volume 1332 of Lecture Notes in Computer Science, pages 25–32. Springer, 1997. 4th European PVM/MPI Users’ Group Meeting.Google Scholar
  7. 7.
    R. Reussner, P. Sanders, L. Prechelt, and M Müller. SKaMPI: A detailed, accurate MPI benchmark. In Vassuk Alexandrov and Jack Dongarra, editors, Recent advances in Parallel Virtual Machine and Message Passing Interface, volume 1497 of Lecture Notes in Computer Science, pages 52–59. Springer, 1998. 5th European PVM/MPI Users’ Group Meeting.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • William Gropp
    • 1
  • Ewing Lusk
    • 1
  1. 1.Argonne National LaboratoryArgonneUSA

Personalised recommendations