Please use this identifier to cite or link to this item: http://dspace.azjhpc.org/xmlui/handle/123456789/36
Full metadata record
DC FieldValueLanguage
dc.contributor.authorAbdulkarimova, Ulviya-
dc.contributor.authorLeonteva, Anna Ouskova-
dc.contributor.authorRolando, Christian-
dc.contributor.authorJeannin-Girardon, Anne-
dc.contributor.authorCollet, Pierre-
dc.date.accessioned2023-04-28T19:49:57Z-
dc.date.available2023-04-28T19:49:57Z-
dc.date.issued2019-12-
dc.identifier.issn2616-6127-
dc.identifier.issn2617-4383-
dc.identifier.otherhttps://doi.org/10.32010/26166127.2019.2.2.122.140-
dc.identifier.urihttp://localhost:8080/xmlui/handle/123456789/36-
dc.description.abstractThis paper describes how transfer-learning can turn a Beowulf cluster into a full super-computer with supra-linear qualitative acceleration. Harmonic Analysis is used as a real-world example to show the kind of result that can be achieved with the proposed supercomputer architecture, that locally exploits absolute space-time parallelism on each machine (SIMD parallelism) and loosely-coupled relative space-time parallelization between different machines (loosely coupled MIMD).en_US
dc.language.isoenen_US
dc.publisherAzerbaijan Journal of High Performance Computingen_US
dc.subjectBeowulf clusteren_US
dc.subjectrelative space-timeen_US
dc.subjectsupra-linear accelerationen_US
dc.subjectqualitative accelerationen_US
dc.subjectGPGPUen_US
dc.subjectloosely coupled machinesen_US
dc.subjectartificial evolutionen_US
dc.subjecttransfer learningen_US
dc.subjectharmonic analysisen_US
dc.subjectsuper-resolutionen_US
dc.subjectnon-uniform samplingen_US
dc.subjectFourier transformen_US
dc.titleTHE PARSEC MACHINE: A NON-NEWTONIAN SUPRA-LINEAR SUPERCOMPUTERen_US
dc.typeArticleen_US
dc.source.journaltitleAzerbaijan Journal of High Performance Computingen_US
dc.source.volume2en_US
dc.source.issue2en_US
dc.source.beginpage122en_US
dc.source.endpage140en_US
dc.source.numberofpages19en_US
Appears in Collections:Azerbaijan Journal of High Performance Computing

Files in This Item:
File Description SizeFormat 
doi.org_10.32010_26166127.2019.2.2.122.140.pdf1.21 MBAdobe PDFView/Open


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