Multi-million particle molecular dynamics. III. Design considerations for data-parallel processing

D. C. Rapaport

doi:10.1016/0010-4655(93)90058-K

Multi-million particle molecular dynamics. III. Design considerations for data-parallel processing

D. C. Rapaport

Department of Physics

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

9 Scopus citations

Abstract

This paper describes an implementation of a parallel molecular dynamics algorithm on the CM2 Connection Machine that is designed for large-scale simulations. The method employs a cell subdivision of the simulation region, and is partly based on the layer approach developed for vector processing. All communication is between adjacent processing elements, eliminating the need for global communication. Performance measurements were made with systems containing over 10⁶ particles.

Original language	English
Pages (from-to)	301-317
Number of pages	17
Journal	Computer Physics Communications
Volume	76
Issue number	3
DOIs	https://doi.org/10.1016/0010-4655(93)90058-K
State	Published - Aug 1993

Bibliographical note

Funding Information:
The author would like to thank the Supercomputer Computations Research Institute at Florida State University for its hospitality while this study was being carried out. The work was supported in part by the US Department of Energy through contract no. DE-FCO5-85ER250000. Hagai Meirovitch and Paul Oppenheimer are thanked for helpful discussion.

Access to Document

10.1016/0010-4655(93)90058-K

Cite this

@article{4d75ccddf8104ee79a1257e005e3a243,

title = "Multi-million particle molecular dynamics. III. Design considerations for data-parallel processing",

abstract = "This paper describes an implementation of a parallel molecular dynamics algorithm on the CM2 Connection Machine that is designed for large-scale simulations. The method employs a cell subdivision of the simulation region, and is partly based on the layer approach developed for vector processing. All communication is between adjacent processing elements, eliminating the need for global communication. Performance measurements were made with systems containing over 106 particles.",

author = "Rapaport, {D. C.}",

note = "Funding Information: The author would like to thank the Supercomputer Computations Research Institute at Florida State University for its hospitality while this study was being carried out. The work was supported in part by the US Department of Energy through contract no. DE-FCO5-85ER250000. Hagai Meirovitch and Paul Oppenheimer are thanked for helpful discussion.",

year = "1993",

month = aug,

doi = "10.1016/0010-4655(93)90058-K",

language = "אנגלית",

volume = "76",

pages = "301--317",

journal = "Computer Physics Communications",

issn = "0010-4655",

publisher = "Elsevier",

number = "3",

}

TY - JOUR

T1 - Multi-million particle molecular dynamics. III. Design considerations for data-parallel processing

AU - Rapaport, D. C.

N1 - Funding Information: The author would like to thank the Supercomputer Computations Research Institute at Florida State University for its hospitality while this study was being carried out. The work was supported in part by the US Department of Energy through contract no. DE-FCO5-85ER250000. Hagai Meirovitch and Paul Oppenheimer are thanked for helpful discussion.

PY - 1993/8

Y1 - 1993/8

N2 - This paper describes an implementation of a parallel molecular dynamics algorithm on the CM2 Connection Machine that is designed for large-scale simulations. The method employs a cell subdivision of the simulation region, and is partly based on the layer approach developed for vector processing. All communication is between adjacent processing elements, eliminating the need for global communication. Performance measurements were made with systems containing over 106 particles.

AB - This paper describes an implementation of a parallel molecular dynamics algorithm on the CM2 Connection Machine that is designed for large-scale simulations. The method employs a cell subdivision of the simulation region, and is partly based on the layer approach developed for vector processing. All communication is between adjacent processing elements, eliminating the need for global communication. Performance measurements were made with systems containing over 106 particles.

UR - http://www.scopus.com/inward/record.url?scp=0027641616&partnerID=8YFLogxK

U2 - 10.1016/0010-4655(93)90058-K

DO - 10.1016/0010-4655(93)90058-K

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:0027641616

SN - 0010-4655

VL - 76

SP - 301

EP - 317

JO - Computer Physics Communications

JF - Computer Physics Communications

IS - 3

ER -

Multi-million particle molecular dynamics. III. Design considerations for data-parallel processing

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this