Skip navigation
Skip navigation

The Design of MPI Based Distributed Shared Memory Systems to Support OpenMP on Clusters

Wong, H'sien; Rendell, Alistair

Description

OpenMP can be supported in cluster environments by using distributed shared memory (DSM) systems. A portable approach for building DSM systems is to layer it on MPI. With these goals in mind, this paper makes two contributions. The first is a discussion about two software DSM systems that we have implemented using MPI. One uses background polling threads while the other uses processes that are driven only by incoming MPI messages. Comparisons of the two approaches show the latter to be a more...[Show more]

dc.contributor.authorWong, H'sien
dc.contributor.authorRendell, Alistair
dc.coverage.spatialAustin USA
dc.date.accessioned2015-12-08T22:38:43Z
dc.date.createdSeptember 19-20 2007
dc.identifier.isbn1424413885
dc.identifier.urihttp://hdl.handle.net/1885/35919
dc.description.abstractOpenMP can be supported in cluster environments by using distributed shared memory (DSM) systems. A portable approach for building DSM systems is to layer it on MPI. With these goals in mind, this paper makes two contributions. The first is a discussion about two software DSM systems that we have implemented using MPI. One uses background polling threads while the other uses processes that are driven only by incoming MPI messages. Comparisons of the two approaches show the latter to be a more scalable architecture that is better suited for the multi-core processors that are becoming commonplace. The second contribution recognizes that a common workaround for sub-team synchronizations in OpenMP is to use the flush directive on shared variables within busy-wait loops. In such a situation, only the flush in the last iteration of the busy-wait loop will result in the conditions necessary for exiting the loop. Thus transfer of the shared value need only be done if there were changes. We implement in our DSM a flush mechanism that eliminates the unnecessary data transfers entirely without any additional support or hints from the programmer.
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE Inc)
dc.relation.ispartofseriesIEEE International Conference on Cluster Computing (Cluster 2007)
dc.sourcePorceedings of IEEE International Conference on Cluster Computing (Cluster 2007)
dc.source.urihttp://www.cluster2007.org/?m_b_c=home
dc.subjectKeywords: Data transfer; Cluster computing; Cluster environments; Distributed shared memory systems; DSM systems; International conferences; Multi-core processors; Scalable architectures; Shared variables; Electric load management
dc.titleThe Design of MPI Based Distributed Shared Memory Systems to Support OpenMP on Clusters
dc.typeConference paper
local.description.notesImported from ARIES
local.description.refereedYes
dc.date.issued2007
local.identifier.absfor080399 - Computer Software not elsewhere classified
local.identifier.ariespublicationU3594520xPUB130
local.type.statusPublished Version
local.contributor.affiliationWong, H'sien, College of Engineering and Computer Science, ANU
local.contributor.affiliationRendell, Alistair, College of Engineering and Computer Science, ANU
local.description.embargo2037-12-31
local.bibliographicCitation.startpage231
local.bibliographicCitation.lastpage240
local.identifier.doi10.1109/CLUSTR.2007.4629236
dc.date.updated2015-12-08T10:10:04Z
local.identifier.scopusID2-s2.0-53349128614
CollectionsANU Research Publications

Download

File Description SizeFormat Image
01_Wong_The_Design_of_MPI_Based_2007.pdf578.34 kBAdobe PDFThumbnail


Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.

Updated:  19 May 2020/ Responsible Officer:  University Librarian/ Page Contact:  Library Systems & Web Coordinator