Advertisement

Block-cyclic array redistribution on networks of workstations

  • Jack Dongarra
  • Frédéric Desprez
  • Antoine Petitet
  • Cyril Randriamaro
  • Yves Robert
5 Algorithms
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1332)

Abstract

This article deals with the run-time comparison between several algorithms (including the MPI_Alltoallv() function call) to perform the redistribution of arrays that are distributed in a block-cyclic fashion over a multidimensional processor grid. The generation of the communication messages to be exchanged by the processors involved in the redistribution is not taken into account. Rather we focus on the scheduling of those messages: how to organize the message exchanges into “structured” communication steps that will minimize communication overhead.

Key-words

distributed arrays redistribution block-cyclic distribution scheduling MPI HPF network of workstations 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Frédéric Desprez, Jack Dongarra, Antoine Petitet, Cyril Randriamaro, and Yves Robert. Scheduling block-cyclic array redistribution. Research Report 97-349, Computer Science Department, University of Tennessee at Knoxville, February 1997. Also available as LAPALK Working Note 120, http://www.netlib.org/lapack/lawns.Google Scholar
  2. 2.
    J. J. Dongarra and D. W. Walker. Software libraries for linear algebra computations on high performance computers. SIAM Review, 37(2):151–180, 1995.CrossRefGoogle Scholar
  3. 3.
    E. T. Kalns and L. M. Ni. Processor mapping techniques towards efficient data redistribution. IEEE Trans. Parallel Distributed Systems, 6(12):1234–1247, 1995.CrossRefGoogle Scholar
  4. 4.
    L. Prylli and B. Tourancheau. Efficient block-cyclic data redistribution. In EuroPar'96, volume 1123 of Lectures Notes in Computer Science, pages 155–164. Springer Verlag, 1996.Google Scholar
  5. 5.
    M. Snir, S. W. Otto, S. Huss-Lederman, D. W. Walker, and J. Dongarra. MPI the complete reference. The MIT Press, 1996.Google Scholar
  6. 6.
    David W. Walker and Steve W.Otto. Redistribution of block-cyclic data distributions using MPI. Concurrency: Practice and Experience, 8(9):707–728, 1996.CrossRefGoogle Scholar
  7. 7.
    Lei Wang, James M. Stichnoth, and Siddhartha Chatterjee. Runtime performance of parallel array assignment: an empirical study. In 1996 ACM/IEEE Supercomputing Conference. http://www.supercomp.org/sc96/proceedings, 1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Jack Dongarra
    • 2
    • 3
  • Frédéric Desprez
    • 1
  • Antoine Petitet
    • 2
  • Cyril Randriamaro
    • 1
  • Yves Robert
    • 2
  1. 1.LIP, Ecole Normale Supérieure de LyonLyon Cedex 07France
  2. 2.Dept. of Computer ScienceUniv. of TennesseeKnoxvilleUSA
  3. 3.Math. Sciences SectionOak Ridge Nat. LaboratoryOak RidgeUSA

Personalised recommendations