Mostrar el registro sencillo del ítem

dc.contributor.authorVilches, Antonio
dc.contributor.authorAsenjo-Plaza, Rafael 
dc.contributor.authorCorbera, Francisco 
dc.contributor.authorNavarro, Ángeles
dc.date.accessioned2014-07-30T10:55:31Z
dc.date.available2014-07-30T10:55:31Z
dc.date.created2014-07-21
dc.date.issued2014-07-30
dc.identifier.urihttp://hdl.handle.net/10630/7956
dc.descriptionEste trabajo describe nuestra contribución para la ejecución de bucles paralelos en arquitecturas multi-core/multi-GPU de forma que la carga computacional se distribuya de forma balanceada entre todas las unidades de computación.es_ES
dc.description.abstractThis paper explores the possibility of efficiently using multicores in conjunction with multiple GPU accelerators under a parallel task programming paradigm. In particular, we address the challenge of extending a parallel_for template to allow its exploitation on heterogeneous systems. The extension is based on a two-stages pipeline engine which is responsible for partitioning and scheduling the chunks into the computational resources. Under this engine, we propose a dynamic scheduling strategy coupled with an adaptive partitioning heuristic that resizes chunks to prevent underutilization and load unbalance of CPUs and GPUs. In this paper we introduce the adaptive partitioning heuristic which is derived from an analytical model that minimizes the load unbalance while maximizes the throughput in the system. Using two benchmarks we evaluate the overhead introduced by our template extensions finding that it is negligible. We also evaluate the efficiency of our adaptive partitioning strategies and compared them with related work.es_ES
dc.description.sponsorshipUniversidad de Málaga. Campus de Excelencia Internacional Andalucía Tech. TIN2010-16144, P08-TIC-3500, P11-TIC-08144es_ES
dc.language.isoenges_ES
dc.rightsinfo:eu-repo/semantics/openAccesses_ES
dc.subjectComputación heterogéneaes_ES
dc.subjectProcesos en paralelo (Informática)es_ES
dc.subject.otherHeterogeneous architecturees_ES
dc.subject.otherDynamic schedulinges_ES
dc.subject.otherAdaptive partitioninges_ES
dc.subject.otherParallel loopes_ES
dc.titleAdaptive Partition Strategies for Loop Parallelism in Heterogeneous Architectureses_ES
dc.typeinfo:eu-repo/semantics/conferenceObjectes_ES
dc.centroE.T.S.I. Informáticaes_ES
dc.relation.eventtitleIntl. Conf. on High Performance Computing and Simulationes_ES
dc.relation.eventplaceBolonia, Italiaes_ES
dc.relation.eventdate21/07/2014es_ES


Ficheros en el ítem

Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem