Parallel molecular dynamics using OpenMP on a shared memory machine
Identifieur interne : 001F01 ( Istex/Curation ); précédent : 001F00; suivant : 001F02Parallel molecular dynamics using OpenMP on a shared memory machine
Auteurs : Raphaël Couturier [France] ; Christophe Chipot [France]Source :
- Computer Physics Communications [ 0010-4655 ] ; 2000.
English descriptors
- Teeft :
- Asymptotic behavior, Available threads, Benchmark applications, Chem, Chipot, Chipot computer physics communications, Communication libraries, Compiler directives, Complex applications, Comput, Computational effort, Couturier, Cpu, Data distribution, Directive, Domain decomposition, Elsevier science, Enddo, Enddo enddo, Enddo enddo enddo, Ewv_local, Ilicon, Intramolecular parameters, Ipoint, Iteration, Jstk mmol, Large number, Limited number, Little knowledge, Master thread, Mber force, Memory multiprocessors, Molecular system, Molecular systems, Multiprocessor workstations, Neutral groups, Newton equations, Optimal performances, Ortran, Osmos, Pairwise interactions, Parallel application, Parallel architectures, Parallel codes, Parallel executable, Parallel machines, Parallel version, Parallelization, Phys, Processor, Raphics, Recherche cnrs, Reduction variables, Respective vapor phase, Rigin, Same code, Same result, Scalar, Scalar code, Scalar codes, Scalar version, Sequential, Sequential program, Simulation, Simulation cell, Subroutine, Temporary array, Third dimension, Vector machines, Wallclock ratios, Water molecules.
Abstract
Abstract: The parallelization of a sequential molecular dynamics (MD) program using OpenMP directives for a Fortran compiler is described. The strength of this approach lies in the possibility to proceed incrementally, with only a few alterations of the original, scalar code. If do-loop splitting with OpenMP is certainly less efficient than domain decomposition (DD) approaches, its fast implementation, requiring little knowledge of the source code, also involves significantly less programming effort. Efficiency of the parallel executable is probed on a distributed-shared memory Silicon Graphics Origin 2000 for MD applications of various complexities. Linear wallclock speed-up ratios using four processors, and ranging between 6.0 and 7.0 with eight, underline the cost-effectiveness of OpenMP directives for fine-grained parallel MD simulations on limited computational resources.
Url:
DOI: 10.1016/S0010-4655(99)00432-4
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: Pour aller vers cette notice dans l'étape Curation :001F26
Links to Exploration step
ISTEX:86F148B8ED2BE762AE60784DE75DFF8DC41C6443Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title>Parallel molecular dynamics using OpenMP on a shared memory machine</title><author><name sortKey="Couturier, Raphael" sort="Couturier, Raphael" uniqKey="Couturier R" first="Raphaël" last="Couturier">Raphaël Couturier</name><affiliation wicri:level="1"><mods:affiliation>Laboratoire Lorrain de Recherche en Informatique et ses Applications, Unité Mixte de Recherche CNRS 7503, 54506 Vandœuvre-lès-Nancy, France</mods:affiliation><country xml:lang="fr">France</country><wicri:regionArea>Laboratoire Lorrain de Recherche en Informatique et ses Applications, Unité Mixte de Recherche CNRS 7503, 54506 Vandœuvre-lès-Nancy</wicri:regionArea></affiliation></author><author><name sortKey="Chipot, Christophe" sort="Chipot, Christophe" uniqKey="Chipot C" first="Christophe" last="Chipot">Christophe Chipot</name><affiliation wicri:level="1"><mods:affiliation>E-mail: chipot@lctn.u-nancy.fr</mods:affiliation><country wicri:rule="url">France</country></affiliation><affiliation wicri:level="1"><mods:affiliation>Laboratoire de Chimie Théorique, Unité Mixte de Recherche CNRS 7565, Université Henri Poincaré, Nancy I, B.P. 239, 54506 Vandœuvre-lès-Nancy, France</mods:affiliation><country xml:lang="fr">France</country><wicri:regionArea>Laboratoire de Chimie Théorique, Unité Mixte de Recherche CNRS 7565, Université Henri Poincaré, Nancy I, B.P. 239, 54506 Vandœuvre-lès-Nancy</wicri:regionArea></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:86F148B8ED2BE762AE60784DE75DFF8DC41C6443</idno><date when="2000" year="2000">2000</date><idno type="doi">10.1016/S0010-4655(99)00432-4</idno><idno type="url">https://api.istex.fr/ark:/67375/6H6-HXNV0GCP-T/fulltext.pdf</idno><idno type="wicri:Area/Istex/Corpus">001F26</idno><idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">001F26</idno><idno type="wicri:Area/Istex/Curation">001F01</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a">Parallel molecular dynamics using OpenMP on a shared memory machine</title><author><name sortKey="Couturier, Raphael" sort="Couturier, Raphael" uniqKey="Couturier R" first="Raphaël" last="Couturier">Raphaël Couturier</name><affiliation wicri:level="1"><mods:affiliation>Laboratoire Lorrain de Recherche en Informatique et ses Applications, Unité Mixte de Recherche CNRS 7503, 54506 Vandœuvre-lès-Nancy, France</mods:affiliation><country xml:lang="fr">France</country><wicri:regionArea>Laboratoire Lorrain de Recherche en Informatique et ses Applications, Unité Mixte de Recherche CNRS 7503, 54506 Vandœuvre-lès-Nancy</wicri:regionArea></affiliation></author><author><name sortKey="Chipot, Christophe" sort="Chipot, Christophe" uniqKey="Chipot C" first="Christophe" last="Chipot">Christophe Chipot</name><affiliation wicri:level="1"><mods:affiliation>E-mail: chipot@lctn.u-nancy.fr</mods:affiliation><country wicri:rule="url">France</country></affiliation><affiliation wicri:level="1"><mods:affiliation>Laboratoire de Chimie Théorique, Unité Mixte de Recherche CNRS 7565, Université Henri Poincaré, Nancy I, B.P. 239, 54506 Vandœuvre-lès-Nancy, France</mods:affiliation><country xml:lang="fr">France</country><wicri:regionArea>Laboratoire de Chimie Théorique, Unité Mixte de Recherche CNRS 7565, Université Henri Poincaré, Nancy I, B.P. 239, 54506 Vandœuvre-lès-Nancy</wicri:regionArea></affiliation></author></analytic><monogr></monogr><series><title level="j">Computer Physics Communications</title><title level="j" type="abbrev">COMPHY</title><idno type="ISSN">0010-4655</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="2000">2000</date><biblScope unit="volume">124</biblScope><biblScope unit="issue">1</biblScope><biblScope unit="page" from="49">49</biblScope><biblScope unit="page" to="59">59</biblScope></imprint><idno type="ISSN">0010-4655</idno></series></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0010-4655</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="Teeft" xml:lang="en"><term>Asymptotic behavior</term><term>Available threads</term><term>Benchmark applications</term><term>Chem</term><term>Chipot</term><term>Chipot computer physics communications</term><term>Communication libraries</term><term>Compiler directives</term><term>Complex applications</term><term>Comput</term><term>Computational effort</term><term>Couturier</term><term>Cpu</term><term>Data distribution</term><term>Directive</term><term>Domain decomposition</term><term>Elsevier science</term><term>Enddo</term><term>Enddo enddo</term><term>Enddo enddo enddo</term><term>Ewv_local</term><term>Ilicon</term><term>Intramolecular parameters</term><term>Ipoint</term><term>Iteration</term><term>Jstk mmol</term><term>Large number</term><term>Limited number</term><term>Little knowledge</term><term>Master thread</term><term>Mber force</term><term>Memory multiprocessors</term><term>Molecular system</term><term>Molecular systems</term><term>Multiprocessor workstations</term><term>Neutral groups</term><term>Newton equations</term><term>Optimal performances</term><term>Ortran</term><term>Osmos</term><term>Pairwise interactions</term><term>Parallel application</term><term>Parallel architectures</term><term>Parallel codes</term><term>Parallel executable</term><term>Parallel machines</term><term>Parallel version</term><term>Parallelization</term><term>Phys</term><term>Processor</term><term>Raphics</term><term>Recherche cnrs</term><term>Reduction variables</term><term>Respective vapor phase</term><term>Rigin</term><term>Same code</term><term>Same result</term><term>Scalar</term><term>Scalar code</term><term>Scalar codes</term><term>Scalar version</term><term>Sequential</term><term>Sequential program</term><term>Simulation</term><term>Simulation cell</term><term>Subroutine</term><term>Temporary array</term><term>Third dimension</term><term>Vector machines</term><term>Wallclock ratios</term><term>Water molecules</term></keywords></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Abstract: The parallelization of a sequential molecular dynamics (MD) program using OpenMP directives for a Fortran compiler is described. The strength of this approach lies in the possibility to proceed incrementally, with only a few alterations of the original, scalar code. If do-loop splitting with OpenMP is certainly less efficient than domain decomposition (DD) approaches, its fast implementation, requiring little knowledge of the source code, also involves significantly less programming effort. Efficiency of the parallel executable is probed on a distributed-shared memory Silicon Graphics Origin 2000 for MD applications of various complexities. Linear wallclock speed-up ratios using four processors, and ranging between 6.0 and 7.0 with eight, underline the cost-effectiveness of OpenMP directives for fine-grained parallel MD simulations on limited computational resources.</div></front></TEI></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Istex/Curation
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001F01 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Curation/biblio.hfd -nk 001F01 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= InforLorV4
|flux= Istex
|étape= Curation
|type= RBID
|clé= ISTEX:86F148B8ED2BE762AE60784DE75DFF8DC41C6443
|texte= Parallel molecular dynamics using OpenMP on a shared memory machine
}}
| This area was generated with Dilib version V0.6.33. |  |