Exploring OpenMP Accelerator Model in a real-life scientific application using hybrid CPU-MIC platforms

Repositorio e-Archivo

Mostrar el registro sencillo del ítem

dc.contributor.author Halbiniak, Kamil
dc.contributor.author Szustak, Lukasz
dc.contributor.author Lastovetsky, Alexey
dc.contributor.author Wyrzykowski, Roman
dc.contributor.editor Carretero Pérez, Jesús
dc.contributor.editor García Blas, Javier
dc.contributor.editor Margenov, Svetozar
dc.contributor.other Universidad Carlos III de Madrid. Computer Architecture, Communications and Systems Group (ARCOS)
dc.date.accessioned 2017-02-17T11:41:52Z
dc.date.available 2017-02-17T11:41:52Z
dc.date.issued 2016-12
dc.identifier.bibliographicCitation Carretero Pérez, Jesús; et.al. (eds.). (2016) Proceedings of the Third International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2016): Sofia, Bulgaria. Universidad Carlos III de Madrid, pp. 11-14
dc.identifier.isbn 978-84-617-7450-0
dc.identifier.uri http://hdl.handle.net/10016/24224
dc.description Proceedings of: Third International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2016). Sofia (Bulgaria), October, 6-7, 2016.
dc.description.abstract The main goal of this paper is the suitability assessment of the OpenMP Accelerator Model (OMPAM) for porting a real-life scientific application to heterogeneous platforms containing a single Intel Xeon Phi coprocessor. This OpenMP extension is supported from version 4.0 of the standard, offering an unified directive-based programming model dedicated for massively parallel accelerators. In our study, we focus on applying the OMPAM extension together with the OpenMP tasks for a parallel application which implements the numerical model of alloy solidification. To map the application efficiently on target hybrid platforms using such constructs as omp target, omp target data and omp target update, we propose a decomposition of main tasks belonging to the computational core of the studied application. In consequence, the coprocessor is used to execute the major parallel workloads, while CPUs are responsible for executing a part of the application that do not require massively parallel resources. Effective overlapping computations with data transfers is another goal achieved in this way. The proposed approach allows us to execute the whole application 3.5 times faster than the original parallel version running on two CPUs.
dc.description.sponsorship This research was conducted with the support of COST Action IC1305 (NESUS), as well as the National Science Centre (Poland) under grant no. UMO-2011/03/B/ST6/03500. The authors are grateful to the Czestochowa University of Technology for granting access to Intel Xeon Phi coprocessors provided by the MICLAB project no. POIG.02.03.00.24-093/13 (http://miclab.pl).
dc.format.extent 4
dc.format.mimetype application/pdf
dc.language.iso eng
dc.relation.ispartof http://hdl.handle.net/10016/24228
dc.subject.other Intel MIC
dc.subject.other Hybrid architecture
dc.subject.other Numerical modeling of solidification
dc.subject.other Heterogeneous programming
dc.subject.other OpenMP Accelerator Model
dc.subject.other Task and data parallelism
dc.title Exploring OpenMP Accelerator Model in a real-life scientific application using hybrid CPU-MIC platforms
dc.type bookPart
dc.type conferenceObject
dc.subject.eciencia Informática
dc.rights.accessRights openAccess
dc.type.version publishedVersion
dc.relation.eventdate October, 6-7, 2016
dc.relation.eventnumber 3
dc.relation.eventplace Sofia, Bulgaria
dc.relation.eventtitle International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2016)
dc.relation.eventtype proceeding
dc.identifier.publicationfirstpage 11
dc.identifier.publicationlastpage 14
dc.identifier.publicationtitle Proceedings of the Third International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2016): Sofia, Bulgaria
 Find Full text

Ficheros en el ítem

*Click en la imagen del fichero para previsualizar.(Los elementos embargados carecen de esta funcionalidad)


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

Mostrar el registro sencillo del ítem