Show simple item record

dc.contributor.authorAliaga Estellés, José Ignacio
dc.contributor.authorBarreda Vayá, María
dc.contributor.authorFlegar, Goran
dc.contributor.authorBollhöffer, Matthias
dc.contributor.authorQuintana Ortí, Enrique S.
dc.date.accessioned2018-01-09T10:29:32Z
dc.date.available2018-01-09T10:29:32Z
dc.date.issued2017-11-10
dc.identifier.citationALIAGA, José I., et al. Communication in task‐parallel ILU‐preconditioned CG solvers using MPI+ OmpSs. Concurrency and Computation: Practice and Experience, 2017, vol. 29, no 21.ca_CA
dc.identifier.issn1532-0626
dc.identifier.issn1532-0634
dc.identifier.urihttp://hdl.handle.net/10234/171616
dc.description.abstractWe target the parallel solution of sparse linear systems via iterative Krylov subspace–based methods enhanced with incomplete LU (ILU)-type preconditioners on clusters of multicore processors. In order to tackle large-scale problems, we develop task-parallel implementations of the classical iteration for the CG method, accelerated via ILUPACK and ILU(0) preconditioners, using MPI + OmpSs. In addition, we integrate several communication-avoiding strategies into the codes, including the butterfly communication scheme and Eijkhout's formulation of the CG method. For all these implementations, we analyze the communication patterns and perform a comparative analysis of their performance and scalability on a cluster consisting of 16 nodes, with 16 cores each.ca_CA
dc.format.extent18 p.ca_CA
dc.format.mimetypeapplication/pdfca_CA
dc.language.isoengca_CA
dc.publisherWileyca_CA
dc.rightsCopyright © John Wiley & Sons, Ltd.ca_CA
dc.subjectILUPACKca_CA
dc.subjectincomplete LU (ILU) preconditionerca_CA
dc.subjectKrylov subspace methodsca_CA
dc.subjectMPIca_CA
dc.subjectOmpSsca_CA
dc.subjectsparse linear systemsca_CA
dc.subjecttask-parallelismca_CA
dc.titleCommunication in task-parallel ILU-preconditioned CG solversusing MPI + OmpSsca_CA
dc.typeinfo:eu-repo/semantics/articleca_CA
dc.identifier.doihttp://dx.doi.org/10.1002/cpe.4280
dc.rights.accessRightsinfo:eu-repo/semantics/openAccessca_CA
dc.relation.publisherVersionConcurrency and Computation: Practice and Experience, 2017, vol. 29, no 21ca_CA
dc.date.embargoEndDate2018-11-10
dc.contributor.funderFEDER / TIN2014-53495-R; Universitat Jaume I / P1-1B2015-26; H2020 EU FETHPC / 671602; Ministerio de Educacion Cultura y Deporte / FPU12/03105; MINECOca_CA
dc.type.versioninfo:eu-repo/semantics/acceptedVersionca_CA


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record