{"id":"https://openalex.org/W4416003993","doi":"https://doi.org/10.1145/3731599.3767539","title":"High-Performance and Power-Efficient Emulation of Matrix Multiplication using INT8 Matrix Engines","display_name":"High-Performance and Power-Efficient Emulation of Matrix Multiplication using INT8 Matrix Engines","publication_year":2025,"publication_date":"2025-11-07","ids":{"openalex":"https://openalex.org/W4416003993","doi":"https://doi.org/10.1145/3731599.3767539"},"language":null,"primary_location":{"id":"doi:10.1145/3731599.3767539","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3731599.3767539","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the SC '25 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3731599.3767539","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5079862429","display_name":"Yuki Uchino","orcid":"https://orcid.org/0000-0002-5906-6624"},"institutions":[{"id":"https://openalex.org/I4210129730","display_name":"RIKEN Center for Computational Science","ror":"https://ror.org/03r519674","country_code":"JP","type":"facility","lineage":["https://openalex.org/I4210110652","https://openalex.org/I4210129730"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Yuki Uchino","raw_affiliation_strings":["RIKEN Center for Computational Science (R-CCS), Kobe, Hyogo, Japan"],"affiliations":[{"raw_affiliation_string":"RIKEN Center for Computational Science (R-CCS), Kobe, Hyogo, Japan","institution_ids":["https://openalex.org/I4210129730"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042608261","display_name":"Katsuhisa Ozaki","orcid":"https://orcid.org/0000-0003-0431-6232"},"institutions":[{"id":"https://openalex.org/I171481255","display_name":"Shibaura Institute of Technology","ror":"https://ror.org/020wjcq07","country_code":"JP","type":"education","lineage":["https://openalex.org/I171481255"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Katsuhisa Ozaki","raw_affiliation_strings":["Shibaura Institute of Technology, Saitama, Saitama, Japan"],"affiliations":[{"raw_affiliation_string":"Shibaura Institute of Technology, Saitama, Saitama, Japan","institution_ids":["https://openalex.org/I171481255"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086152822","display_name":"Toshiyuki Imamura","orcid":"https://orcid.org/0000-0003-1601-9710"},"institutions":[{"id":"https://openalex.org/I4210129730","display_name":"RIKEN Center for Computational Science","ror":"https://ror.org/03r519674","country_code":"JP","type":"facility","lineage":["https://openalex.org/I4210110652","https://openalex.org/I4210129730"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Toshiyuki Imamura","raw_affiliation_strings":["RIKEN Center for Computational Science (R-CCS), Kobe, Hyogo, Japan"],"affiliations":[{"raw_affiliation_string":"RIKEN Center for Computational Science (R-CCS), Kobe, Hyogo, Japan","institution_ids":["https://openalex.org/I4210129730"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5079862429"],"corresponding_institution_ids":["https://openalex.org/I4210129730"],"apc_list":null,"apc_paid":null,"fwci":4.3072,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.94694467,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1824","last_page":"1831"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.43479999899864197,"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":0.43479999899864197,"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/T11697","display_name":"Numerical Methods and Algorithms","score":0.12780000269412994,"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"}},{"id":"https://openalex.org/T10363","display_name":"Low-power high-performance VLSI design","score":0.11230000108480453,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/emulation","display_name":"Emulation","score":0.9696999788284302},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.8543000221252441},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.628000020980835},{"id":"https://openalex.org/keywords/matrix","display_name":"Matrix (chemical analysis)","score":0.5741999745368958},{"id":"https://openalex.org/keywords/power","display_name":"Power (physics)","score":0.5605999827384949},{"id":"https://openalex.org/keywords/hardware-emulation","display_name":"Hardware emulation","score":0.5105999708175659}],"concepts":[{"id":"https://openalex.org/C149810388","wikidata":"https://www.wikidata.org/wiki/Q5374873","display_name":"Emulation","level":2,"score":0.9696999788284302},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8543000221252441},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6782000064849854},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.628000020980835},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.5741999745368958},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.5605999827384949},{"id":"https://openalex.org/C94115699","wikidata":"https://www.wikidata.org/wiki/Q5656406","display_name":"Hardware emulation","level":3,"score":0.5105999708175659},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.5033000111579895},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.38339999318122864},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.38179999589920044},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32600000500679016},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.29660001397132874},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.28850001096725464},{"id":"https://openalex.org/C147297375","wikidata":"https://www.wikidata.org/wiki/Q6674930","display_name":"Look-ahead","level":2,"score":0.2630000114440918}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3731599.3767539","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3731599.3767539","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the SC '25 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2508.03984","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.03984","pdf_url":"https://arxiv.org/pdf/2508.03984","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.1145/3731599.3767539","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3731599.3767539","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the SC '25 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G4382947259","display_name":null,"funder_award_id":"23K28100, 24K23874, 25H01109, 25K03126","funder_id":"https://openalex.org/F4320334764","funder_display_name":"Japan Society for the Promotion of Science"}],"funders":[{"id":"https://openalex.org/F4320334764","display_name":"Japan Society for the Promotion of Science","ror":"https://ror.org/00hhkn466"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":8,"referenced_works":["https://openalex.org/W2073327480","https://openalex.org/W2079232854","https://openalex.org/W2169631286","https://openalex.org/W2981980078","https://openalex.org/W4221160294","https://openalex.org/W4392884779","https://openalex.org/W4405129106","https://openalex.org/W4406214914"],"related_works":[],"abstract_inverted_index":{"Recent":[0],"architectures":[1],"integrate":[2],"high-performance":[3],"and":[4,31,37,73,97,128],"power-efficient":[5],"matrix":[6,15,44],"engines.":[7,45],"These":[8],"engines":[9],"demonstrate":[10],"remarkable":[11],"performance":[12,127],"in":[13,20,77,101],"low-precision":[14,43],"multiplication,":[16],"which":[17],"is":[18],"crucial":[19],"deep":[21],"learning.":[22],"Several":[23],"techniques":[24],"have":[25],"been":[26],"proposed":[27,65,89,119],"to":[28,81,105,114],"emulate":[29],"single-":[30],"double-precision":[32],"general":[33],"matrix-matrix":[34],"multiplication":[35],"(SGEMM":[36],"DGEMM,":[38],"respectively)":[39],"by":[40],"leveraging":[41],"such":[42],"In":[46],"this":[47],"study,":[48],"we":[49],"present":[50],"emulation":[51,67,91,116,120],"methods":[52],"that":[53],"significantly":[54],"outperforms":[55],"conventional":[56,115],"approaches.":[57],"On":[58],"a":[59,69,74,93,98],"GH200":[60],"Grace":[61],"Hopper":[62],"Superchip,":[63],"the":[64,118],"DGEMM":[66,83],"achieves":[68,92,121],"1.4":[70],"\u00d7":[71,95,125],"speedup":[72,96],"43%":[75],"improvement":[76,100],"power":[78,102,130],"efficiency":[79,103],"compared":[80,104,113],"native":[82,106],"for":[84,108],"sufficiently":[85,109],"large":[86,110],"problems.":[87,111],"The":[88],"SGEMM":[90,107],"3.0":[94],"154%":[99],"Furthermore,":[112],"methods,":[117],"more":[122],"than":[123],"2":[124],"higher":[126],"superior":[129],"efficiency.":[131]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-04-15T08:11:43.952461","created_date":"2025-11-07T00:00:00"}