Publication:
Improving the Performance of the MPI_Allreduce Collective Operation through Rank Renaming

Simple item page
dc.affiliation.dptoUC3M. Departamento de Informáticaes
dc.affiliation.grupoinvUC3M. Grupo de Investigación: Arquitectura de Computadores, Comunicaciones y Sistemases
dc.contributor.authorRico-Gallego, Juan-Antonio
dc.contributor.authorDíaz-Martín, Juan-Carlos
dc.contributor.editorCarretero Pérez, Jesús
dc.contributor.editorGarcía Blas, Javier
dc.contributor.editorBarbosa, Jorge
dc.contributor.editorMorla, Ricardo
dc.contributor.otherUniversidad Carlos III de Madrid. Computer Architecture, Communications and Systems Group (ARCOS)
dc.date.accessioned2015-11-11T10:30:47Z
dc.date.available2015-11-11T10:30:47Z
dc.date.issued2014-11
dc.descriptionProceedings of: First International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2014). Porto (Portugal), August 27-28, 2014.en
dc.description.abstractCollective operations, a key issue in the global efficiency of HPC applications, are optimized in current MPI libraries by choosing at runtime between a set of algorithms, based on platform-dependent beforehand established parameters, as the message size or the number of processes. However, with progressively more cores per node, the cost of a collective algorithm must be mainly imputed to process-to-processor mapping, because its decisive influence over the network traffic. Hierarchical design of collective algorithms pursuits to minimize the data movement through the slowest communication channels of the multi-core cluster. Nevertheless, the hierarchical implementation of some collectives becomes inefficient, and even impracticable, due to the operation definition itself. This paper proposes a new approach that departs from a frequently found regular mapping, either sequential or round-robin. While keeping the mapping, the rank assignation to the processes is temporarily changed prior to the execution of the collective algorithm. The new assignation makes the communication pattern to adapt to the communication channels hierarchy. We explore this technique for the Ring algorithm when used in the well-known MPI_Allreduce collective, and discuss the obtained performance results. Extensions to other algorithms and collective operations are proposed.en
dc.description.sponsorshipThe work presented in this paper has been partially supported by EU under the COST programme Action IC1305, ’Network for Sustainable Ultrascale Computing (NESUS)’, and by the computing facilities of Extremadura Research Centre for Advanced Technologies (CETACIEMAT), funded by the European Regional Development Fund (ERDF). CETA-CIEMAT belongs to CIEMAT and the Government of Spain.en
dc.format.extent6
dc.format.mimetypeapplication/pdf
dc.identifier.bibliographicCitationCarretero Pérez, Jesús; et.al. (eds.). (2014) Proceedings of the First International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2014): Porto, Portugal. Universidad Carlos III de Madrid, pp. 1-6.en
dc.identifier.isbn978-84-617-2251-8
dc.identifier.publicationfirstpage1
dc.identifier.publicationlastpage6
dc.identifier.publicationtitleProceedings of the First International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2014): Porto, Portugalen
dc.identifier.urihttps://hdl.handle.net/10016/21978
dc.language.isoeng
dc.relation.eventdateAugust 27-28, 2014en
dc.relation.eventnumber1
dc.relation.eventplacePorto, Portugalen
dc.relation.eventtitleInternational Workshop on Sustainable Ultrascale Computing Systems (NESUS 2014)en
dc.rights.accessRightsopen access
dc.subject.ecienciaInformáticaes
dc.subject.otherMPI Collectivesen
dc.subject.otherParallel algorithmsen
dc.subject.otherMessage passing interfaceen
dc.subject.otherMulti-core clustersen
dc.titleImproving the Performance of the MPI_Allreduce Collective Operation through Rank Renamingen
dc.typeconference paper*
dc.type.hasVersionVoR*
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
improving_NESUS_2014.pdf
Size:
551.92 KB
Format:
Adobe Portable Document Format
Download
Collections
First International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2014)