乐胖代购免代理版

{"id":"https://openalex.org/W2964017942","doi":"https://doi.org/10.1109/ipdps.2019.00022","title":"Fast Batched Matrix Multiplication for Small Sizes Using Half-Precision Arithmetic on GPUs","display_name":"Fast Batched Matrix Multiplication for Small Sizes Using Half-Precision Arithmetic on GPUs","publication_year":2019,"publication_date":"2019-05-01","ids":{"openalex":"https://openalex.org/W2964017942","doi":"https://doi.org/10.1109/ipdps.2019.00022","mag":"2964017942"},"language":"en","primary_location":{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/ipdps.2019.00022","pdf_url":null,"source":{"id":"https://openalex.org/S4363607067","display_name":"2022 IEEE International Parallel and Distributed Processing Symposium (IPDPS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false},"type":"article","type_crossref":"proceedings-article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5077268543","display_name":"Ahmad Abdelfattah","orcid":"https://orcid.org/0000-0001-5054-4784"},"institutions":[{"id":"https://openalex.org/I75027704","display_name":"University of Tennessee at Knoxville","ror":"https://ror.org/020f3ap87","country_code":"US","type":"education","lineage":["https://openalex.org/I75027704"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ahmad Abdelfattah","raw_affiliation_strings":["Innovative Computing Laboratory, University of Tennessee, USA"],"affiliations":[{"raw_affiliation_string":"Innovative Computing Laboratory, University of Tennessee, USA","institution_ids":["https://openalex.org/I75027704"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083604741","display_name":"Stanimire Tomov","orcid":"https://orcid.org/0000-0002-5937-7959"},"institutions":[{"id":"https://openalex.org/I75027704","display_name":"University of Tennessee at Knoxville","ror":"https://ror.org/020f3ap87","country_code":"US","type":"education","lineage":["https://openalex.org/I75027704"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Stanimire Tomov","raw_affiliation_strings":["Innovative Computing Laboratory, University of Tennessee, USA"],"affiliations":[{"raw_affiliation_string":"Innovative Computing Laboratory, University of Tennessee, USA","institution_ids":["https://openalex.org/I75027704"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5075517045","display_name":"Jack Dongarra","orcid":"https://orcid.org/0000-0003-3247-1782"},"institutions":[{"id":"https://openalex.org/I1289243028","display_name":"Oak Ridge National Laboratory","ror":"https://ror.org/01qz5mb56","country_code":"US","type":"facility","lineage":["https://openalex.org/I1289243028","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I4210159294"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jack Dongarra","raw_affiliation_strings":["Oak Ridge National Laboratory, USA"],"affiliations":[{"raw_affiliation_string":"Oak Ridge National Laboratory, USA","institution_ids":["https://openalex.org/I1289243028"]}]}],"institution_assertions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":3.219,"has_fulltext":false,"cited_by_count":34,"citation_normalized_percentile":{"value":0.999156,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"111","last_page":"122"},"is_retracted":false,"is_paratext":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9999,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9999,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12303","display_name":"Tensor decomposition and applications","score":0.9948,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10792","display_name":"Matrix Theory and Algorithms","score":0.9886,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/linear-algebra","display_name":"Linear algebra","score":0.6989474},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.6330129},{"id":"https://openalex.org/keywords/double-precision-floating-point-format","display_name":"Double-precision floating-point format","score":0.47163293},{"id":"https://openalex.org/keywords/matrix","display_name":"Matrix (chemical analysis)","score":0.42797118},{"id":"https://openalex.org/keywords/single-precision-floating-point-format","display_name":"Single-precision floating-point format","score":0.42776436}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.78411925},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.7575594},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.74199563},{"id":"https://openalex.org/C139352143","wikidata":"https://www.wikidata.org/wiki/Q82571","display_name":"Linear algebra","level":2,"score":0.6989474},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6635461},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.6330129},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.5992039},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.525779},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.49377784},{"id":"https://openalex.org/C35912277","wikidata":"https://www.wikidata.org/wiki/Q1243369","display_name":"Double-precision floating-point format","level":3,"score":0.47163293},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.42797118},{"id":"https://openalex.org/C133095886","wikidata":"https://www.wikidata.org/wiki/Q1307173","display_name":"Single-precision floating-point format","level":3,"score":0.42776436},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.419187},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.29176062},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1283246},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.12562913},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.09143138},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"is_oa":false,"landing_page_url":"https://doi.org/10.1109/ipdps.2019.00022","pdf_url":null,"source":{"id":"https://openalex.org/S4363607067","display_name":"2022 IEEE International Parallel and Distributed Processing Symposium (IPDPS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.41,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"grants":[],"datasets":[],"versions":[],"referenced_works_count":34,"referenced_works":["https://openalex.org/W1004371317","https://openalex.org/W1487564550","https://openalex.org/W1530262073","https://openalex.org/W1667652561","https://openalex.org/W1841592590","https://openalex.org/W1863336885","https://openalex.org/W1978642402","https://openalex.org/W2002555321","https://openalex.org/W2063186542","https://openalex.org/W2090593986","https://openalex.org/W2091371833","https://openalex.org/W2093843662","https://openalex.org/W2099021415","https://openalex.org/W2138215414","https://openalex.org/W2155893237","https://openalex.org/W2162322364","https://openalex.org/W2172654076","https://openalex.org/W2183156981","https://openalex.org/W2268122618","https://openalex.org/W2499931820","https://openalex.org/W2531348778","https://openalex.org/W2544527082","https://openalex.org/W2751100001","https://openalex.org/W2767612671","https://openalex.org/W2781998146","https://openalex.org/W2808102735","https://openalex.org/W2887740241","https://openalex.org/W2888025781","https://openalex.org/W2895305554","https://openalex.org/W2963374099","https://openalex.org/W2966371448","https://openalex.org/W3141650078","https://openalex.org/W4298857548","https://openalex.org/W4302296459"],"related_works":["https://openalex.org/W4302308373","https://openalex.org/W4302004660","https://openalex.org/W4288634132","https://openalex.org/W4283812690","https://openalex.org/W3150370983","https://openalex.org/W3145384893","https://openalex.org/W3022016791","https://openalex.org/W2763768774","https://openalex.org/W2239119680","https://openalex.org/W1638830944"],"abstract_inverted_index":{"Matrix":[0],"multiplication":[1],"(GEMM)":[2],"is":[3,14,19,158],"the":[4,39,72,83,117,136,144,148,151,171,174,193,201,206,231],"most":[5,32],"important":[6],"operation":[7,17,85],"in":[8,21,52,76,125,138],"dense":[9,56],"linear":[10,43,57],"algebra.":[11],"Because":[12],"it":[13],"a":[15,109,159,166,223],"computebound":[16],"that":[18,113,121,142,164,189,200],"rich":[20],"data":[22],"reuse,":[23],"many":[24,87],"applications":[25],"from":[26],"different":[27],"scientific":[28],"domains":[29,53],"cast":[30],"their":[31],"performancecritical":[33],"stages":[34],"to":[35,170,183,214],"use":[36],"GEMM.":[37],"With":[38],"rise":[40],"of":[41,86,116,153,168,185],"batch":[42],"algebra,":[44],"batched":[45,74,96],"GEMM":[46,75,97],"operations":[47],"have":[48],"become":[49],"increasingly":[50],"popular":[51],"other":[54],"than":[55],"solvers,":[58,65],"such":[59],"as":[60],"tensor":[61],"contractions,":[62],"sparse":[63],"direct":[64],"and":[66,220,237],"machine":[67],"learning.":[68],"In":[69],"particular":[70,181],"for":[71,98,211],"latter,":[73],"reduced":[77],"precision":[78],"(i.e.,":[79],"FP16)":[80],"has":[81],"been":[82],"core":[84],"deep":[88],"learning":[89],"frameworks.":[90],"This":[91],"paper":[92,178],"introduces":[93],"an":[94,139],"optimized":[95,140,208],"FP16":[99],"arithmetic":[100],"(HGEMM)":[101],"using":[102,222],"graphics":[103],"processing":[104],"units":[105],"(GPUs).":[106],"We":[107],"provide":[108],"detailed":[110],"design":[111,141,203],"strategy":[112],"takes":[114],"advantage":[115],"Tensor":[118,194],"Core":[119,195],"technology":[120],"was":[122],"recently":[123],"introduced":[124],"CUDA-enabled":[126],"GPUs.":[127],"The":[128,156,177],"developed":[129],"solution":[130],"uses":[131],"low-level":[132],"APIs":[133],"provided":[134],"by":[135,147,216],"vendor":[137,209],"overcomes":[143],"limitations":[145],"imposed":[146],"hardware":[149],"(in":[150],"form":[152],"discrete":[154],"configurations).":[155],"outcome":[157],"highly":[160,207],"flexible":[161],"GPU":[162],"kernel":[163],"provides":[165],"lot":[167],"controls":[169],"developer,":[172],"despite":[173],"aforementioned":[175],"restrictions.":[176],"also":[179],"pays":[180],"attention":[182],"multiplications":[184],"very":[186],"small":[187,229],"matrices":[188],"cannot":[190],"fully":[191],"occupy":[192],"units.":[196],"Our":[197],"results":[198],"show":[199],"proposed":[202],"can":[204],"outperform":[205],"routine":[210],"sizes":[212],"up":[213],"100":[215],"factors":[217],"between":[218,235],"1.2\u00d7":[219],"10\u00d7":[221],"Tesla":[224],"V100":[225],"GPU.":[226],"For":[227],"extremely":[228],"matrices,":[230],"observed":[232],"speedups":[233],"range":[234],"1.8\u00d7":[236],"26\u00d7.":[238]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W2964017942","counts_by_year":[{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":7},{"year":2021,"cited_by_count":9},{"year":2020,"cited_by_count":9},{"year":2019,"cited_by_count":4}],"updated_date":"2024-12-07T15:58:49.860009","created_date":"2019-07-30"}