Show simple item record

dc.contributor.authorPfister, Hanspeter
dc.contributor.authorEdgar, Richard G
dc.contributor.authorMitchell, Daniel A
dc.contributor.authorOrd, Stephen M
dc.contributor.authorGreenhill, Lincoln J.
dc.contributor.authorClark, Michael A.
dc.contributor.authorDale, Kevin
dc.contributor.authorWayth, Randall B.
dc.date.accessioned2011-11-10T20:00:38Z
dc.date.issued2010
dc.identifier.citationEdgara, Richard G., Michael A. Clark, Kevin Dale, Daniel A. Mitchell, Stephen M. Ord, Randall B. Wayth, Hanspeter Pfister, and Lincoln J. Greenhill. 2010. Enabling a high throughput real time data pipeline for a large radio telescope array with GPUs. Computer Physics Communications 181(10): 1707-1714.en_US
dc.identifier.issn0010-4655en_US
dc.identifier.urihttp://nrs.harvard.edu/urn-3:HUL.InstRepos:5344187
dc.description.abstractThe Murchison Widefield Array (MWA) is a next-generation radio telescope currently under construction in the remote Western Australia Outback. Raw data will be generated continuously at 5 GiB s\(^{−1}\), grouped into 8 s cadences. This high throughput motivates the development of on-site, real time processing and reduction in preference to archiving, transport and off-line processing. Each batch of 8 s data must be completely reduced before the next batch arrives. Maintaining real time operation will require a sustained performance of around 2.5 TFLOP s\(^{−1}\) (including convolutions, FFTs, interpolations and matrix multiplications). We describe a scalable heterogeneous computing pipeline implementation, exploiting both the high computing density and FLOP-per-Watt ratio of modern GPUs. The architecture is highly parallel within and across nodes, with all major processing elements performed by GPUs. Necessary scatter-gather operations along the pipeline are loosely synchronized between the nodes hosting the GPUs. The MWA will be a frontier scientific instrument and a pathfinder for planned peta- and exascale facilities.en_US
dc.description.sponsorshipEngineering and Applied Sciencesen_US
dc.language.isoen_USen_US
dc.publisherElsevieren_US
dc.relation.isversionofdoi:10.1016/j.cpc.2010.06.019en_US
dc.relation.hasversionhttp://gvi.seas.harvard.edu/sites/all/files/CPC_2010.pdfen_US
dash.licenseOAP
dc.subjectradio telescopes and instrumentationen_US
dc.subjectheterodyne receiversen_US
dc.subjectmathematical procedures and computer techniquesen_US
dc.subjectcomputer science and technologyen_US
dc.titleEnabling a High Throughput Real Time Data Pipeline for a Large Radio Telescope Array with GPUsen_US
dc.typeJournal Articleen_US
dc.description.versionAccepted Manuscripten_US
dc.relation.journalComputer Physics Communicationsen_US
dash.depositing.authorPfister, Hanspeter
dc.date.available2011-11-10T20:00:38Z
dash.affiliation.otherInitiative in Innovative Computingen_US
dc.identifier.doi10.1016/j.cpc.2010.06.019*
dash.authorsorderedfalse
dash.contributor.affiliatedGreenhill, Lincoln
dash.contributor.affiliatedMitchell, Daniel
dash.contributor.affiliatedEdgar, Richard G
dash.contributor.affiliatedClark, Michael A.
dash.contributor.affiliatedOrd, Stephen
dash.contributor.affiliatedPfister, Hanspeter


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record