Research Article

Redistribution strategies for portable parallel FFT: a case study

  1. Anshu Dubey1,*,
  2. Daniele Tessera2

Article first published online: 21 MAR 2001

DOI: 10.1002/cpe.564

Issue

Concurrency and Computation: Practice and Experience

Concurrency and Computation: Practice and Experience

Volume 13, Issue 3, pages 209–220, March 2001

Additional Information

How to Cite

Dubey, A. and Tessera, D. (2001), Redistribution strategies for portable parallel FFT: a case study. Concurrency and Computation: Practice and Experience, 13: 209–220. doi: 10.1002/cpe.564

Author Information

  1. 1

    Astronomy & Astrophysics, University of Chicago, 5640 S. Ellis Ave., Chicago, IL 60637, U.S.A.

  2. 2

    Dipartimento di Informatica e Sistemistica, Università degli Studi di Pavia, I-27100 Pavia, Italy

*Astronomy & Astrophysics, University of Chicago, 5640 S. Ellis Ave., Chicago, IL 60637, U.S.A.

Publication History

  1. Issue published online: 21 MAR 2001
  2. Article first published online: 21 MAR 2001
  3. Manuscript Revised: 23 NOV 2000
  4. Manuscript Received: 10 APR 2000

SEARCH

SEARCH BY CITATION

ARTICLE TOOLS

Get PDF (325K)

Keywords:

  • FFT;
  • distributed transpose;
  • MPI; communication policy

Abstract

The best approach to parallelize multidimensional FFT algorithms has long been under debate. Distributed transposes are widely used, but they also vary in communication policies and hence performance. In this work we analyze the impact of different redistribution strategies on the performance of parallel FFT, on various machine architectures. We found that some redistribution strategies were consistently superior, while some others were unexpectedly inferior. An in-depth investigation into the reasons for this behavior is included in this work. Copyright © 2001 John Wiley & Sons, Ltd.

Get PDF (325K)