Chapter

Recent Advances in Parallel Virtual Machine and Message Passing Interface

Volume 4192 of the series Lecture Notes in Computer Science pp 40-48

MPI Collective Algorithm Selection and Quadtree Encoding

  • Jelena Pješivac–GrbovićAffiliated withInnovative Computing Laboratory, University of Tennessee Computer Science Department
  • , Graham E. FaggAffiliated withInnovative Computing Laboratory, University of Tennessee Computer Science Department
  • , Thara AngskunAffiliated withInnovative Computing Laboratory, University of Tennessee Computer Science Department
  • , George BosilcaAffiliated withInnovative Computing Laboratory, University of Tennessee Computer Science Department
  • , Jack J. DongarraAffiliated withInnovative Computing Laboratory, University of Tennessee Computer Science Department

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Selecting the close-to-optimal collective algorithm based on the parameters of the collective call at run time is an important step in achieving good performance of MPI applications. In this paper, we focus on MPI collective algorithm selection process and explore the applicability of the quadtree encoding method to this problem. We construct quadtrees with different properties from the measured algorithm performance data and analyze the quality and performance of decision functions generated from these trees. The experimental data shows that in some cases, the decision function based on a quadtree structure with a mean depth of 3 can incur as little as a 5% performance penalty on average. The exact, experimentally measured, decision function for all tested collectives could be fully represented using quadtrees with a maximum of 6 levels. These results indicate that quadtrees may be a feasible choice for both processing of the performance data and automatic decision function generation.