Towards a Better Expressiveness of the Speedup Metric in MPI Context

Jean Baptiste Besnard, Allen D. Malony, Sameer Shende, Marc Perache, Patrick Carribault, Julien Jaeger

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Many-core processors are imposing new constraints to parallel applications. In particular, the MPI+X model or hybridization is becoming a compulsory avenue to extract performance by mitigating both memory and communication overhead. In this context, performance tools also have to evolve in order to represent more complex states combining multiple runtimes and programming models. In this paper, we propose to start from a well-known performance metric, the Speedup, showing that it can be bounded by the acceleration of any program section. From this observation, we propose a compact tool-oriented MPI abstraction providing such time slices (or phases). We demonstrate the benefits of this approach first on a simple benchmark, identifying factors limiting speedup. And second, using an MPI+OpenMP benchmark to measure OpenMP scaling solely from MPI instrumentation.

Original languageEnglish
Title of host publicationProceedings - 46th International Conference on Parallel Processing Workshops, ICPPW 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages251-260
Number of pages10
ISBN (Electronic)9781538610442
DOIs
Publication statusPublished - 5 Sept 2017
Externally publishedYes
Event46th International Conference on Parallel Processing Workshops, ICPPW 2017 - Bristol, United Kingdom
Duration: 14 Aug 2017 → …

Publication series

NameProceedings of the International Conference on Parallel Processing Workshops
ISSN (Print)1530-2016

Conference

Conference46th International Conference on Parallel Processing Workshops, ICPPW 2017
Country/TerritoryUnited Kingdom
CityBristol
Period14/08/17 → …

Keywords

  • MPI
  • MPI-Section
  • Phase
  • Profiling
  • Speedup

Fingerprint

Dive into the research topics of 'Towards a Better Expressiveness of the Speedup Metric in MPI Context'. Together they form a unique fingerprint.

Cite this