Hierarchical collectives in MPICH2

Hao Zhu, David Goodell, William Gropp, Rajeev Thakur

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Most parallel systems on which MPI is used are now hierarchical, such as systems with SMP nodes. Many papers have shown algorithms that exploit shared memory to optimize collective operations to good effect. But how much of the performance benefit comes from tailoring the algorithm to the hierarchical topology of the system? We describe an implementation of many of the MPI collectives based entirely on message-passing primitives that exploits the two-level hierarchy. Our results show that exploiting shared memory directly usually gives small additional benefit and suggests design approaches for where the benefit is large.

Original languageEnglish (US)
Title of host publicationRecent Advances in Parallel Virtual Machine and Message Passing Interface - 16th European PVM/MPI Users' Group Meeting, Proceedings
PublisherSpringer
Pages325-326
Number of pages2
ISBN (Print)3642037690, 9783642037696
DOIs
StatePublished - 2009
Event16th European Parallel Virtual Machine and Message Passing Interface Users' Group Meeting, EuroPVM/MPI - Espoo, Finland
Duration: Sep 7 2009Sep 10 2009

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5759 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other16th European Parallel Virtual Machine and Message Passing Interface Users' Group Meeting, EuroPVM/MPI
Country/TerritoryFinland
CityEspoo
Period9/7/099/10/09

Keywords

  • Collective Communication
  • MPI

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Hierarchical collectives in MPICH2'. Together they form a unique fingerprint.

Cite this