Implementation and evaluation of shared-memory communication and synchronization operations in MPICH2 using the Nemesis communication subsystem

Darius Buntinas; Guillaume Mercier; William Gropp

doi:10.1016/j.parco.2007.06.003

Implementation and evaluation of shared-memory communication and synchronization operations in MPICH2 using the Nemesis communication subsystem

Darius Buntinas, Guillaume Mercier, William Gropp

Research output: Contribution to journal › Article › peer-review

Abstract

This paper presents the implementation of MPICH2 over the Nemesis communication subsystem and the evaluation of its shared-memory performance. We describe design issues as well as some of the optimization techniques we employed. We conducted a performance evaluation over shared memory using microbenchmarks. The evaluation shows that MPICH2 Nemesis has very low communication overhead, making it suitable for smaller-grained applications.

Original language	English (US)
Pages (from-to)	634-644
Number of pages	11
Journal	Parallel Computing
Volume	33
Issue number	9
DOIs	https://doi.org/10.1016/j.parco.2007.06.003
State	Published - Sep 2007
Externally published	Yes

Keywords

Communication subsystem
MPI
MPI implementation
Message passing
Parallel computing
Shared-memory

ASJC Scopus subject areas

Software
Theoretical Computer Science
Hardware and Architecture
Computer Networks and Communications
Computer Graphics and Computer-Aided Design
Artificial Intelligence

Online availability

10.1016/j.parco.2007.06.003

Library availability

Discover UIUC Full Text

Cite this

@article{6ae78686e120487a9c08ece22eb0222b,

title = "Implementation and evaluation of shared-memory communication and synchronization operations in MPICH2 using the Nemesis communication subsystem",

abstract = "This paper presents the implementation of MPICH2 over the Nemesis communication subsystem and the evaluation of its shared-memory performance. We describe design issues as well as some of the optimization techniques we employed. We conducted a performance evaluation over shared memory using microbenchmarks. The evaluation shows that MPICH2 Nemesis has very low communication overhead, making it suitable for smaller-grained applications.",

keywords = "Communication subsystem, MPI, MPI implementation, Message passing, Parallel computing, Shared-memory",

author = "Darius Buntinas and Guillaume Mercier and William Gropp",

note = "Funding Information: This work was supported by the Mathematical, Information, and Computational Sciences Division subprogram of the Office of Advanced Scientific Computing Research, Office of Science, US Department of Energy, under Contract DE-AC02-06CH11357. ",

year = "2007",

month = sep,

doi = "10.1016/j.parco.2007.06.003",

language = "English (US)",

volume = "33",

pages = "634--644",

journal = "Parallel Computing",

issn = "0167-8191",

publisher = "Elsevier B.V.",

number = "9",

}

TY - JOUR

T1 - Implementation and evaluation of shared-memory communication and synchronization operations in MPICH2 using the Nemesis communication subsystem

AU - Buntinas, Darius

AU - Mercier, Guillaume

AU - Gropp, William

N1 - Funding Information: This work was supported by the Mathematical, Information, and Computational Sciences Division subprogram of the Office of Advanced Scientific Computing Research, Office of Science, US Department of Energy, under Contract DE-AC02-06CH11357.

PY - 2007/9

Y1 - 2007/9

N2 - This paper presents the implementation of MPICH2 over the Nemesis communication subsystem and the evaluation of its shared-memory performance. We describe design issues as well as some of the optimization techniques we employed. We conducted a performance evaluation over shared memory using microbenchmarks. The evaluation shows that MPICH2 Nemesis has very low communication overhead, making it suitable for smaller-grained applications.

AB - This paper presents the implementation of MPICH2 over the Nemesis communication subsystem and the evaluation of its shared-memory performance. We describe design issues as well as some of the optimization techniques we employed. We conducted a performance evaluation over shared memory using microbenchmarks. The evaluation shows that MPICH2 Nemesis has very low communication overhead, making it suitable for smaller-grained applications.

KW - Communication subsystem

KW - MPI

KW - MPI implementation

KW - Message passing

KW - Parallel computing

KW - Shared-memory

UR - http://www.scopus.com/inward/record.url?scp=34548689127&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34548689127&partnerID=8YFLogxK

U2 - 10.1016/j.parco.2007.06.003

DO - 10.1016/j.parco.2007.06.003

M3 - Article

AN - SCOPUS:34548689127

SN - 0167-8191

VL - 33

SP - 634

EP - 644

JO - Parallel Computing

JF - Parallel Computing

IS - 9

ER -

Implementation and evaluation of shared-memory communication and synchronization operations in MPICH2 using the Nemesis communication subsystem

Abstract

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this