000845262 001__ 845262
000845262 005__ 20210129233336.0
000845262 0247_ $$2doi$$a10.1007/s11227-017-2232-2
000845262 0247_ $$2ISSN$$a0920-8542
000845262 0247_ $$2ISSN$$a1573-0484
000845262 0247_ $$2Handle$$a2128/18243
000845262 0247_ $$2WOS$$aWOS:000428284000010
000845262 037__ $$aFZJ-2018-02543
000845262 082__ $$a004
000845262 1001_ $$0P:(DE-Juel1)132124$$aHalver, Rene$$b0
000845262 245__ $$aFunction portability of molecular dynamics on heterogeneous parallel architectures with OpenCL
000845262 260__ $$aDordrecht [u.a.]$$bSpringer Science + Business Media B.V$$c2018
000845262 3367_ $$2DRIVER$$aarticle
000845262 3367_ $$2DataCite$$aOutput Types/Journal article
000845262 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1524722920_8481
000845262 3367_ $$2BibTeX$$aARTICLE
000845262 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000845262 3367_ $$00$$2EndNote$$aJournal Article
000845262 520__ $$aClassical molecular dynamics simulation for atomistic systems is implemented in OpenCL and benchmarked on a variety of different hardware platforms. Modifying the number of particles and system size in the study provides insight into characteristics of parallel compute platforms, where latency, data transfer, memory access characteristics and compute intense work can be identified as fingerprints in benchmark runs. Data layouts are compared, for which the access of structure-of-arrays shows best performance in most cases. It is demonstrated that function portability can be achieved straightforwardly with OpenCL, while performance portability lacks behind as various architectures strongly depend on specific vectorisation optimisation.
000845262 536__ $$0G:(DE-HGF)POF3-511$$a511 - Computational Science and Mathematical Methods (POF3-511)$$cPOF3-511$$fPOF III$$x0
000845262 588__ $$aDataset connected to CrossRef
000845262 7001_ $$0P:(DE-Juel1)132142$$aHomberg, Wilhelm$$b1
000845262 7001_ $$0P:(DE-Juel1)132274$$aSutmann, Godehard$$b2$$eCorresponding author
000845262 773__ $$0PERI:(DE-600)1479917-0$$a10.1007/s11227-017-2232-2$$gVol. 74, no. 4, p. 1522 - 1533$$n4$$p1522 - 1533$$tThe journal of supercomputing$$v74$$x1573-0484$$y2018
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/10.1007_s11227-017-2232-2.pdf$$yRestricted
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/10.1007_s11227-017-2232-2.gif?subformat=icon$$xicon$$yRestricted
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/10.1007_s11227-017-2232-2.jpg?subformat=icon-1440$$xicon-1440$$yRestricted
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/10.1007_s11227-017-2232-2.jpg?subformat=icon-180$$xicon-180$$yRestricted
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/10.1007_s11227-017-2232-2.jpg?subformat=icon-640$$xicon-640$$yRestricted
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/10.1007_s11227-017-2232-2.pdf?subformat=pdfa$$xpdfa$$yRestricted
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/FuncPortMDOpenCL.pdf$$yPublished on 2018-01-23. Available in OpenAccess from 2019-01-23.
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/FuncPortMDOpenCL.gif?subformat=icon$$xicon$$yPublished on 2018-01-23. Available in OpenAccess from 2019-01-23.
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/FuncPortMDOpenCL.jpg?subformat=icon-1440$$xicon-1440$$yPublished on 2018-01-23. Available in OpenAccess from 2019-01-23.
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/FuncPortMDOpenCL.jpg?subformat=icon-180$$xicon-180$$yPublished on 2018-01-23. Available in OpenAccess from 2019-01-23.
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/FuncPortMDOpenCL.jpg?subformat=icon-640$$xicon-640$$yPublished on 2018-01-23. Available in OpenAccess from 2019-01-23.
000845262 8564_ $$uhttps://juser.fz-juelich.de/record/845262/files/FuncPortMDOpenCL.pdf?subformat=pdfa$$xpdfa$$yPublished on 2018-01-23. Available in OpenAccess from 2019-01-23.
000845262 909CO $$ooai:juser.fz-juelich.de:845262$$pdnbdelivery$$pdriver$$pVDB$$popen_access$$popenaire
000845262 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)132124$$aForschungszentrum Jülich$$b0$$kFZJ
000845262 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)132142$$aForschungszentrum Jülich$$b1$$kFZJ
000845262 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)132274$$aForschungszentrum Jülich$$b2$$kFZJ
000845262 9131_ $$0G:(DE-HGF)POF3-511$$1G:(DE-HGF)POF3-510$$2G:(DE-HGF)POF3-500$$3G:(DE-HGF)POF3$$4G:(DE-HGF)POF$$aDE-HGF$$bKey Technologies$$lSupercomputing & Big Data$$vComputational Science and Mathematical Methods$$x0
000845262 9141_ $$y2018
000845262 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS
000845262 915__ $$0StatID:(DE-HGF)1160$$2StatID$$aDBCoverage$$bCurrent Contents - Engineering, Computing and Technology
000845262 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search
000845262 915__ $$0StatID:(DE-HGF)0530$$2StatID$$aEmbargoed OpenAccess
000845262 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bJ SUPERCOMPUT : 2015
000845262 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection
000845262 915__ $$0StatID:(DE-HGF)0110$$2StatID$$aWoS$$bScience Citation Index
000845262 915__ $$0StatID:(DE-HGF)0111$$2StatID$$aWoS$$bScience Citation Index Expanded
000845262 915__ $$0StatID:(DE-HGF)9900$$2StatID$$aIF < 5
000845262 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC
000845262 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline
000845262 915__ $$0StatID:(DE-HGF)0420$$2StatID$$aNationallizenz
000845262 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bThomson Reuters Master Journal List
000845262 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0
000845262 980__ $$ajournal
000845262 980__ $$aVDB
000845262 980__ $$aI:(DE-Juel1)JSC-20090406
000845262 980__ $$aUNRESTRICTED
000845262 9801_ $$aFullTexts