Massively parallel implementation of a fast multipole method for distributed memory machines

Jakub Kurzak; B. Montgomery Pettitt

doi:10.1016/j.jpdc.2005.02.001

Massively parallel implementation of a fast multipole method for distributed memory machines

Jakub Kurzak, B. Montgomery Pettitt

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

32 Scopus citations

Abstract

We present a new load balanced parallel implementation of a non-adaptive version of Greengard and Rokhlin's fast multipole method for distributed memory architectures with focus on applications in molecular dynamics. We introduce a novel load balancing and communication overlapping scheme. Our implementation includes periodic boundary conditions calculations and facilitates multiple time stepping techniques without sacrificing determinism of computation and scales to hundreds of processor for systems of only O(10k) atoms.

Original language	English (US)
Pages (from-to)	870-881
Number of pages	12
Journal	Journal of Parallel and Distributed Computing
Volume	65
Issue number	7
DOIs	https://doi.org/10.1016/j.jpdc.2005.02.001
State	Published - Jul 2005
Externally published	Yes

Keywords

Molecular dynamics algorithms

ASJC Scopus subject areas

Software
Theoretical Computer Science
Hardware and Architecture
Computer Networks and Communications
Artificial Intelligence

Access to Document

10.1016/j.jpdc.2005.02.001

Cite this

@article{af9a5f96edcd44448164f613c5820c15,

title = "Massively parallel implementation of a fast multipole method for distributed memory machines",

abstract = "We present a new load balanced parallel implementation of a non-adaptive version of Greengard and Rokhlin's fast multipole method for distributed memory architectures with focus on applications in molecular dynamics. We introduce a novel load balancing and communication overlapping scheme. Our implementation includes periodic boundary conditions calculations and facilitates multiple time stepping techniques without sacrificing determinism of computation and scales to hundreds of processor for systems of only O(10k) atoms.",

keywords = "Molecular dynamics algorithms",

author = "Jakub Kurzak and Pettitt, {B. Montgomery}",

note = "Funding Information: The authors thank the NIH, NASA and the R. A. Welch Foundation for partial support of their work. The calculations were made possible through generous computing time allocations of the Pittsburgh Supercomputing Center, Pacific Northwest National Laboratory, Texas Learning and Computation Center and High Performance Computing Center of the University of Houston. A portion of the research described in this paper was performed as part of a Computational Grand Challenge Project in the Environmental Molecular Sciences Laboratory, sponsored by the Department of Energy{\textquoteright}s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory.",

year = "2005",

month = jul,

doi = "10.1016/j.jpdc.2005.02.001",

language = "English (US)",

volume = "65",

pages = "870--881",

journal = "Journal of Parallel and Distributed Computing",

issn = "0743-7315",

publisher = "Academic Press Inc.",

number = "7",

}

TY - JOUR

T1 - Massively parallel implementation of a fast multipole method for distributed memory machines

AU - Kurzak, Jakub

AU - Pettitt, B. Montgomery

N1 - Funding Information: The authors thank the NIH, NASA and the R. A. Welch Foundation for partial support of their work. The calculations were made possible through generous computing time allocations of the Pittsburgh Supercomputing Center, Pacific Northwest National Laboratory, Texas Learning and Computation Center and High Performance Computing Center of the University of Houston. A portion of the research described in this paper was performed as part of a Computational Grand Challenge Project in the Environmental Molecular Sciences Laboratory, sponsored by the Department of Energy’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory.

PY - 2005/7

Y1 - 2005/7

N2 - We present a new load balanced parallel implementation of a non-adaptive version of Greengard and Rokhlin's fast multipole method for distributed memory architectures with focus on applications in molecular dynamics. We introduce a novel load balancing and communication overlapping scheme. Our implementation includes periodic boundary conditions calculations and facilitates multiple time stepping techniques without sacrificing determinism of computation and scales to hundreds of processor for systems of only O(10k) atoms.

AB - We present a new load balanced parallel implementation of a non-adaptive version of Greengard and Rokhlin's fast multipole method for distributed memory architectures with focus on applications in molecular dynamics. We introduce a novel load balancing and communication overlapping scheme. Our implementation includes periodic boundary conditions calculations and facilitates multiple time stepping techniques without sacrificing determinism of computation and scales to hundreds of processor for systems of only O(10k) atoms.

KW - Molecular dynamics algorithms

UR - http://www.scopus.com/inward/record.url?scp=19944419779&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=19944419779&partnerID=8YFLogxK

U2 - 10.1016/j.jpdc.2005.02.001

DO - 10.1016/j.jpdc.2005.02.001

M3 - Article

AN - SCOPUS:19944419779

SN - 0743-7315

VL - 65

SP - 870

EP - 881

JO - Journal of Parallel and Distributed Computing

JF - Journal of Parallel and Distributed Computing

IS - 7

ER -

Massively parallel implementation of a fast multipole method for distributed memory machines

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this