Evaluating the Performance of MPI-2 Dynamic Communicators and One-Sided Communication

  • Edgar Gabriel
  • Graham E. Fagg
  • Jack J. Dongarra
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2840)

Abstract

This paper evaluates the performance of several MPI implementations regarding two chapters of the MPI-2 specification. First, we analyze, whether the performance using dynamically created communicators is comparable to the approach presented in MPI-1 using a static communicator for different MPI libraries. We than evaluate, whether the communication performance of one-sided communication on current machines, represents a benefit or a drawback to the end-user compared to the more conventional two-sided communication.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Gropp, W., Lusk, E., Doss, N., Skjellum, A.: A high-performance, portable implementation of the MPI message-passing interface standard. Parallel Computing 22 (1996)Google Scholar
  2. 2.
    Message Passing Interface Forum. MPI: A Message-Passing Interface Standard (version 1.1). Technical report (June 1995), http://www.mpi-forum.org
  3. 3.
    Message Passing Interface Forum. MPI-2: Extentions to the Message-Passing Interface, July 18 (1997)Google Scholar
  4. 4.
    Rabenseifner, R., Koniges, A.E.: Effective File-I/O Bandwidth Benchmark. In: Bode, A., Ludwig, T., Karl, W.C., Wismüller, R. (eds.) Euro-Par 2000. LNCS, vol. 1900, pp. 1273–1283. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  5. 5.
    Golebiewski, M., Traeff, J.L.: MPI-2 One-Sided Communications in a Giganet SMP Cluster. In: Cotronis, Y., Dongarra, J. (eds.) Recent Advances in Parallel Virtual Machine and Message Passing Interface, pp. 16–23. Springer, Heidelberg (2001)CrossRefGoogle Scholar
  6. 6.
    Luecke, G., Hu, W.: Evaluating the Performance of MPI-2 One-Sided Routines on a Cray SV-1, technical report, December 21 (2002), http://www.public.iastete.edu/grl/publications.html
  7. 7.
    Booth, S., Mourao, E.: Single Sided MPI implementations for SUN MPI. In: Proceedings of Supercomputing 2000, Dallas, TX, USA (2000)Google Scholar
  8. 8.
    Mierendorff, H., Cassirer, K., Schwamborn, H.: Working with MPI Benchmark Suites on ccNUMA Architectures. In: Dongarra, J., Kacsuk, P., Podhorszki, N. (eds.) Recent Advances in Parallel Virtual Machine and Message Passing Interface, pp. 18–26. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  9. 9.
    Gropp, W., Lusk, E.: Reproducible measurements of MPI performance characteristics. In: Dongarra, J.J., Luque, E., Margalef, T. (eds.) Recent advances in Parallel Virtual Vachine and Message Passing Interface, pp. 11–18. Springer, Heidelberg (1999)CrossRefGoogle Scholar
  10. 10.
    Reussner, R.H., Sanders, P., Prechelt, L., Müller, M.: SKaMPI: A Detailed Accurate MPI Benchmark. In: Alexandrov, V., Dongarra, J.J. (eds.) Recent advances in Parallel Virtual Vachine and Message Passing Interface, pp. 52–59. Springer, Heidelberg (1999)Google Scholar
  11. 11.
    Hempel, R.: Basic message passing benchmarks, methodology and pitfalls. Presented at SPEC Workshop, Wuppertal, Germany (September 1999)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Edgar Gabriel
    • 1
  • Graham E. Fagg
    • 1
  • Jack J. Dongarra
    • 1
  1. 1.Innovative Computing Laboratory, Computer Science DepartementUniversity of TennesseeKnoxvilleUSA

Personalised recommendations