MPI Overlap: Benchmark and Analysis

Abstract : In HPC applications, one of the major overhead compared to sequentiel code, is communication cost. Application programmers often amortize this cost by overlapping communications with computation. To do so, they post a non-blocking MPI request, perform computation, and wait for communication completion, assuming MPI communication will progress in background. In this paper, we propose to measure what really happens when trying to overlap non-blocking point-to-point communications with computation. We explain how background progression works, we describe relevant test cases, we identify challenges for a benchmark, then we propose a benchmark suite to measure how much overlap happen in various cases. We exhibit overlap benchmark results on a wide panel of MPI libraries and hardware platforms. Finally, we classify, analyze, and explain the results using low-level traces to reveal the internal behavior of the MPI library.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [19 references]  Display  Hide  Download

https://hal.inria.fr/hal-01324179
Contributor : Alexandre Denis <>
Submitted on : Tuesday, May 31, 2016 - 4:51:46 PM
Last modification on : Thursday, February 7, 2019 - 2:48:42 PM
Document(s) archivé(s) le : Thursday, September 1, 2016 - 11:38:27 AM

File

final.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01324179, version 1

Citation

Alexandre Denis, François Trahay. MPI Overlap: Benchmark and Analysis. International Conference on Parallel Processing, Aug 2016, Philadelphia, United States. ⟨hal-01324179⟩

Share

Metrics

Record views

540

Files downloads

1024