{"id":"https://openalex.org/W4413146721","doi":"https://doi.org/10.1109/ipdpsw66978.2025.00135","title":"Millions of Matrix-Multiplications: GEMM Variations on Aurora","display_name":"Millions of Matrix-Multiplications: GEMM Variations on Aurora","publication_year":2025,"publication_date":"2025-06-03","ids":{"openalex":"https://openalex.org/W4413146721","doi":"https://doi.org/10.1109/ipdpsw66978.2025.00135"},"language":"en","primary_location":{"id":"doi:10.1109/ipdpsw66978.2025.00135","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ipdpsw66978.2025.00135","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5017529089","display_name":"Colleen Bertoni","orcid":"https://orcid.org/0009-0004-0552-0061"},"institutions":[{"id":"https://openalex.org/I1282105669","display_name":"Argonne National Laboratory","ror":"https://ror.org/05gvnxz63","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282105669","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Colleen Bertoni","raw_affiliation_strings":["Argonne Leadership Computing Facility Argonne National Laboratory,Lemont,IL,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Argonne Leadership Computing Facility Argonne National Laboratory,Lemont,IL,USA","institution_ids":["https://openalex.org/I1282105669"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047914322","display_name":"Thomas Applencourt","orcid":"https://orcid.org/0000-0001-7522-9449"},"institutions":[{"id":"https://openalex.org/I1282105669","display_name":"Argonne National Laboratory","ror":"https://ror.org/05gvnxz63","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282105669","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Thomas Applencourt","raw_affiliation_strings":["Argonne Leadership Computing Facility Argonne National Laboratory,Lemont,IL,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Argonne Leadership Computing Facility Argonne National Laboratory,Lemont,IL,USA","institution_ids":["https://openalex.org/I1282105669"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061322208","display_name":"Longfei Gao","orcid":"https://orcid.org/0000-0002-7790-7538"},"institutions":[{"id":"https://openalex.org/I1282105669","display_name":"Argonne National Laboratory","ror":"https://ror.org/05gvnxz63","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282105669","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Longfei Gao","raw_affiliation_strings":["Argonne Leadership Computing Facility Argonne National Laboratory,Lemont,IL,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Argonne Leadership Computing Facility Argonne National Laboratory,Lemont,IL,USA","institution_ids":["https://openalex.org/I1282105669"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5026603316","display_name":"Ti Leggett","orcid":"https://orcid.org/0000-0003-2899-1603"},"institutions":[{"id":"https://openalex.org/I1282105669","display_name":"Argonne National Laboratory","ror":"https://ror.org/05gvnxz63","country_code":"US","type":"facility","lineage":["https://openalex.org/I1282105669","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I40347166"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ti Leggett","raw_affiliation_strings":["Argonne Leadership Computing Facility Argonne National Laboratory,Lemont,IL,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Argonne Leadership Computing Facility Argonne National Laboratory,Lemont,IL,USA","institution_ids":["https://openalex.org/I1282105669"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I1282105669"],"apc_list":null,"apc_paid":null,"fwci":20.0526,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.99053628,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"850","last_page":"856"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12504","display_name":"Mathematics and Applications","score":0.7156999707221985,"subfield":{"id":"https://openalex.org/subfields/2608","display_name":"Geometry and Topology"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12504","display_name":"Mathematics and Applications","score":0.7156999707221985,"subfield":{"id":"https://openalex.org/subfields/2608","display_name":"Geometry and Topology"},"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":0.6748999953269958,"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/computer-science","display_name":"Computer science","score":0.593727707862854},{"id":"https://openalex.org/keywords/matrix","display_name":"Matrix (chemical analysis)","score":0.4584573805332184},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.42129766941070557},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.37603670358657837},{"id":"https://openalex.org/keywords/computer-graphics","display_name":"Computer graphics (images)","score":0.3550637364387512},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.32347339391708374}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.593727707862854},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.4584573805332184},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.42129766941070557},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.37603670358657837},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.3550637364387512},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.32347339391708374},{"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/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ipdpsw66978.2025.00135","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ipdpsw66978.2025.00135","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.5400000214576721,"display_name":"Climate action","id":"https://metadata.un.org/sdg/13"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306084","display_name":"U.S. Department of Energy","ror":"https://ror.org/01bj3aw27"},{"id":"https://openalex.org/F4320337506","display_name":"Advanced Scientific Computing Research","ror":"https://ror.org/0012c7r22"},{"id":"https://openalex.org/F4320338284","display_name":"Argonne National Laboratory","ror":"https://ror.org/05gvnxz63"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2051487156","https://openalex.org/W2073681303","https://openalex.org/W4409439182","https://openalex.org/W2053286651","https://openalex.org/W2181743346","https://openalex.org/W2187401768","https://openalex.org/W2181413294","https://openalex.org/W2052122378","https://openalex.org/W2544423928","https://openalex.org/W2062023542"],"abstract_inverted_index":{"Supercomputers":[0],"are":[1,106,177],"typically":[2],"composed":[3],"of":[4,8,10,22,51,54,64,68],"hundreds":[5],"to":[6,44,153,179],"tens":[7],"thousands":[9],"nodes.":[11],"Although":[12,171],"one":[13,21],"may":[14],"benchmark":[15],"and":[16,82,87,107,130,148,155,169,175],"optimize":[17],"kernels":[18],"on":[19,66,71,79,84,186],"just":[20],"those":[23],"nodes,":[24],"there":[25,141],"will":[26],"be":[27,184],"hardware":[28],"or":[29],"manufacturing":[30],"differences":[31],"between":[32,120],"each":[33,67,80,85],"node,":[34,86],"which":[35],"can":[36],"impact":[37],"overall":[38],"performance.":[39],"Hence,":[40],"it":[41],"is":[42,122],"important":[43],"understand":[45],"the":[46,49,58,62,72,102,118,172,181],"variation":[47,119],"in":[48],"performance":[50,63,76],"individual":[52],"components":[53,105,121,147],"a":[55,112,133,151,163],"supercomputer.To":[56],"investigate":[57],"variation,":[59],"we":[60,149],"measured":[61,78],"GEMMs":[65],"9762":[69],"nodes":[70],"supercomputer":[73,166],"Aurora.":[74],"The":[75],"was":[77],"CPU":[81],"GPU":[83],"with":[88],"multiple":[89],"data":[90],"types":[91,127],"(FP64,":[92],"FP32,":[93],"TF32,":[94],"BF16,":[95],"I8).":[96],"We":[97,115,137],"evaluated":[98],"how":[99,108],"spread":[100],"out":[101],"measurements":[103],"across":[104],"well":[109],"they":[110],"match":[111],"Gaussian":[113,134],"distribution.":[114],"find":[116,139],"that":[117,140],"low":[123],"for":[124,165],"all":[125],"GEMM":[126],"(\u2264":[128],"3.0%)":[129],"most":[131],"fit":[132],"distribution":[135],"well.":[136],"also":[138],"were":[142],"some":[143],"significantly":[144],"poorly":[145],"performing":[146],"discuss":[150],"method":[152],"identify":[154],"remove":[156],"these":[157,160],"outliers.We":[158],"provide":[159],"results":[161],"as":[162],"reference":[164],"users,":[167],"benchmarkers,":[168],"maintainers.":[170],"exact":[173],"variations":[174],"distributions":[176],"specific":[178],"Aurora,":[180],"methods":[182],"could":[183],"repeated":[185],"any":[187],"large":[188],"system.":[189]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}
