Show simple item record

dc.contributor.authorCastelló, Adrián
dc.contributor.authorPena, Antonio J.
dc.contributor.authorMayo, Rafael
dc.contributor.authorPlanas, Judit
dc.contributor.authorQuintana Ortí, Enrique S.
dc.contributor.authorBalaji, Pavan
dc.date.accessioned2016-07-13T09:22:47Z
dc.date.available2016-07-13T09:22:47Z
dc.date.issued2016-06-21
dc.identifier.citationCASTELLÓ, Adrián, et al. Exploring the interoperability of remote GPGPU virtualization using rCUDA and directive-based programming models. The Journal of Supercomputing, 2016, p. 1-15ca_CA
dc.identifier.issn0920-8542
dc.identifier.urihttp://hdl.handle.net/10234/161666
dc.description.abstractDirective-based programming models, such as OpenMP, OpenACC, and OmpSs, enable users to accelerate applications by using coprocessors with little effort. These devices offer significant computing power, but their use can introduce two problems: an increase in the total cost of ownership and their underutilization because not all codes match their architecture. Remote accelerator virtualization frameworks address those problems. In particular, rCUDA provides transparent access to any graphic processor unit installed in a cluster, reducing the number of accelerators and increasing their utilization ratio. Joining these two technologies, directive-based programming models and rCUDA, is thus highly appealing. In this work, we study the integration of OmpSs and OpenACC with rCUDA, describing and analyzing several applications over three different hardware configurations that include two InfiniBand interconnections and three NVIDIA accelerators. Our evaluation reveals favorable performance results, showing low overhead and similar scaling factors when using remote accelerators instead of local devices.ca_CA
dc.description.sponsorShipThe researchers from the Universitat Jaume I de Castell o were supported by Universitat Jaume I research project (P11B2013-21), project TIN2014-53495-R, a Generalitat Valenciana grant and FEDER. The researcher from the Barcelona Supercomputing Center (BSC-CNS) was supported by the European Commission (HiPEAC-3 Network of Excellence, FP7-ICT 287759), Intel-BSC Exascale Lab collaboration, IBM/BSC Exascale Initiative collaboration agreement, Computaci on de Altas Prestaciones VI (TIN2012-34557) and the Generalitat de Catalunya (2014-SGR-1051). This work was partially supported by the U.S. Dept. of Energy, O ce of Science, O ce of Advanced Scienti c Computing Research (SC-21), under contract DE-AC02-06CH11357. The initial version of rCUDA was jointly developed by Universitat Polit ecnica de Val encia (UPV) and Universitat Jaume I de Castell on (UJI) until year 2010. This initial development was later split into two branches. Part of the UPV version was used in this paper. The development of the UPV branch was supported by Generalitat Valenciana under Grants PROMETEO 2008/060 and Prometeo II 2013/009. We gratefully acknowledge the computing resources provided and operated by the Joint Laboratory for System Evaluation (JLSE) at Argonne National Laboratory.ca_CA
dc.format.extent15 p.ca_CA
dc.format.mimetypeapplication/pdfca_CA
dc.language.isoengca_CA
dc.publisherSpringerca_CA
dc.relation.isFormatOfhttp://link.springer.com/article/10.1007/s11227-016-1791-yca_CA
dc.relation.isPartOfThe Journal of Supercomputing, 2016ca_CA
dc.rights© 2016 Springer International Publishing. Part of Springer Nature.ca_CA
dc.subjectGPUsca_CA
dc.subjectDirective-based programming modelsca_CA
dc.subjectOpenACCca_CA
dc.subjectOmpSsca_CA
dc.subjectRemote virtualizationca_CA
dc.subjectrCUDAca_CA
dc.titleExploring the interoperability of remote GPGPU virtualization using rCUDA and directive-based programming modelsca_CA
dc.typeinfo:eu-repo/semantics/articleca_CA
dc.identifier.doihttp://dx.doi.org/10.1007/s11227-016-1791-y
dc.rights.accessRightsinfo:eu-repo/semantics/openAccessca_CA
dc.relation.publisherVersionhttp://link.springer.com/article/10.1007/s11227-016-1791-yca_CA


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record