Vis enkel innførsel

dc.contributor.authorNaim, Md.
dc.contributor.authorManne, Fredrik
dc.contributor.authorHalappanavar, Mahantesh
dc.contributor.authorTumeo, Antonino
dc.date.accessioned2017-10-13T12:07:59Z
dc.date.available2017-10-13T12:07:59Z
dc.date.issued2017
dc.identifier.urihttps://hdl.handle.net/1956/16753
dc.description.abstractWe present and evaluate a new GPU algorithm based on the Louvain method for community detection. Our algorithm is the first for this problem that parallelizes the access to individual edges. In this way we can fine tune the load balance when processing networks with nodes of highly varying degrees. This is achieved by scaling the number of threads assigned to each node according to its degree. Extensive experiments show that we obtain speedups up to a factor of 270 compared to the sequential algorithm. The algorithm consistently outperforms other recent shared memory implementations and is only one order of magnitude slower than the current fastest parallel Louvain method running on a Blue Gene/Q supercomputer using more than 500K threads.en_US
dc.language.isoengeng
dc.publisherIEEEen_US
dc.relation.ispartof<a href="http://hdl.handle.net/1956/16755" target="_blank">Parallel Matching and Clustering Algorithms on GPUs</a>en_US
dc.relation.ispartof2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS)
dc.titleCommunity Detection on the GPUen_US
dc.typeChapter
dc.typePeer reviewed
dc.description.versionacceptedVersionen_US
dc.rights.holderCopyright 2017 IEEEen_US
dc.identifier.doihttps://doi.org/10.1109/ipdps.2017.16
dc.source.pagenumber625-634
dc.identifier.citationIn: Parallel and Distributed Processing Symposium (IPDPS), 2017 IEEE International (pp. 625-634)


Tilhørende fil(er)

Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel