{"id":"https://openalex.org/W7128909541","doi":"https://doi.org/10.2197/ipsjjip.34.132","title":"Single-precision Matrix Multiplication Performance on Cerebras CS-2: Evaluation and Modelling of Performance, Scalability and Energy Efficiency","display_name":"Single-precision Matrix Multiplication Performance on Cerebras CS-2: Evaluation and Modelling of Performance, Scalability and Energy Efficiency","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7128909541","doi":"https://doi.org/10.2197/ipsjjip.34.132"},"language":"en","primary_location":{"id":"doi:10.2197/ipsjjip.34.132","is_oa":true,"landing_page_url":"https://doi.org/10.2197/ipsjjip.34.132","pdf_url":"https://www.jstage.jst.go.jp/article/ipsjjip/34/0/34_132/_pdf","source":{"id":"https://openalex.org/S4210239267","display_name":"Journal of Information Processing","issn_l":"1882-6652","issn":["1882-6652"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Information Processing","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://www.jstage.jst.go.jp/article/ipsjjip/34/0/34_132/_pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022019173","display_name":"Takaaki Miyajima","orcid":"https://orcid.org/0000-0001-9409-0913"},"institutions":[{"id":"https://openalex.org/I16656306","display_name":"Meiji University","ror":"https://ror.org/02rqvrp93","country_code":"JP","type":"education","lineage":["https://openalex.org/I16656306"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Takaaki Miyajima","raw_affiliation_strings":["Meiji University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meiji University","institution_ids":["https://openalex.org/I16656306"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5146629985","display_name":"Ryunosuke Matsuzaki","orcid":null},"institutions":[{"id":"https://openalex.org/I16656306","display_name":"Meiji University","ror":"https://ror.org/02rqvrp93","country_code":"JP","type":"education","lineage":["https://openalex.org/I16656306"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Ryunosuke Matsuzaki","raw_affiliation_strings":["Meiji University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Meiji University","institution_ids":["https://openalex.org/I16656306"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5035972892","display_name":"Daichi Mukunoki","orcid":"https://orcid.org/0000-0002-0051-6811"},"institutions":[{"id":"https://openalex.org/I60134161","display_name":"Nagoya University","ror":"https://ror.org/04chrp450","country_code":"JP","type":"education","lineage":["https://openalex.org/I60134161"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Daichi Mukunoki","raw_affiliation_strings":["Information Technology Center, Nagoya University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Information Technology Center, Nagoya University","institution_ids":["https://openalex.org/I60134161"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.09757528,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"34","issue":"0","first_page":"132","last_page":"139"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.7831000089645386,"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":0.7831000089645386,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.057500001043081284,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.02199999988079071,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/scalability","display_name":"Scalability","score":0.7570000290870667},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.7186999917030334},{"id":"https://openalex.org/keywords/multiplication","display_name":"Multiplication (music)","score":0.6312000155448914},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.5482000112533569},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.5475999712944031},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.5349000096321106},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.47850000858306885}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8490999937057495},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7570000290870667},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.7186999917030334},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.6312000155448914},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5482000112533569},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.5475999712944031},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.5349000096321106},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.47850000858306885},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.45210000872612},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.42890000343322754},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4178999960422516},{"id":"https://openalex.org/C42747912","wikidata":"https://www.wikidata.org/wiki/Q1048447","display_name":"Multiplicative function","level":2,"score":0.40230000019073486},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.3790999948978424},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3637999892234802},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.36059999465942383},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.35760000348091125},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3497999906539917},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.29820001125335693},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.27549999952316284},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.26159998774528503}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.2197/ipsjjip.34.132","is_oa":true,"landing_page_url":"https://doi.org/10.2197/ipsjjip.34.132","pdf_url":"https://www.jstage.jst.go.jp/article/ipsjjip/34/0/34_132/_pdf","source":{"id":"https://openalex.org/S4210239267","display_name":"Journal of Information Processing","issn_l":"1882-6652","issn":["1882-6652"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Information Processing","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.2197/ipsjjip.34.132","is_oa":true,"landing_page_url":"https://doi.org/10.2197/ipsjjip.34.132","pdf_url":"https://www.jstage.jst.go.jp/article/ipsjjip/34/0/34_132/_pdf","source":{"id":"https://openalex.org/S4210239267","display_name":"Journal of Information Processing","issn_l":"1882-6652","issn":["1882-6652"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Information Processing","raw_type":"journal-article"},"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.9049139022827148}],"awards":[{"id":"https://openalex.org/G2919449371","display_name":"Performance estimation of scientific computing of an extremely large AI accelerator with 850,000 PEs.","funder_award_id":"24K14972","funder_id":"https://openalex.org/F4320334764","funder_display_name":"Japan Society for the Promotion of Science"},{"id":"https://openalex.org/G7724793919","display_name":"\u6b21\u4e16\u4ee3\u306e\u9ad8\u52b9\u7387\u8a08\u7b97\u57fa\u76e4\u3092\u5b9f\u73fe\u3059\u308b\u9069\u5fdc\u578b\u30c7\u30fc\u30bf\u5727\u7e2e\u30cf\u30fc\u30c9\u30a6\u30a7\u30a2\u306e\u63a2\u6c42","funder_award_id":"JPMJAP2341","funder_id":"https://openalex.org/F4320334789","funder_display_name":"Japan Science and Technology Agency"}],"funders":[{"id":"https://openalex.org/F4320334764","display_name":"Japan Society for the Promotion of Science","ror":"https://ror.org/00hhkn466"},{"id":"https://openalex.org/F4320334789","display_name":"Japan Science and Technology Agency","ror":"https://ror.org/00097mb19"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7128909541.pdf","grobid_xml":"https://content.openalex.org/works/W7128909541.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W164687769","https://openalex.org/W201315547","https://openalex.org/W2168748809","https://openalex.org/W2169631286","https://openalex.org/W2808362339","https://openalex.org/W2984305089","https://openalex.org/W3133446571","https://openalex.org/W4250730856","https://openalex.org/W4297097318","https://openalex.org/W4321636661","https://openalex.org/W4381327281","https://openalex.org/W4406157123"],"related_works":[],"abstract_inverted_index":{"Although":[0],"recent":[1],"supercomputers":[2],"have":[3,30],"been":[4,31],"improving":[5],"their":[6],"computational":[7],"performance,":[8,107,112],"achieving":[9],"performance":[10,55,116,129,138,157],"scaling":[11,111,146],"with":[12,68,89],"respect":[13],"to":[14,23,33],"the":[15,54,69,73,104,123,127,160],"number":[16],"of":[17,56,130,141,148],"nodes":[18],"is":[19,48,62],"not":[20],"easy":[21],"due":[22],"long":[24],"inter-node":[25],"communication":[26,35],"latency.":[27],"Many":[28],"attempts":[29],"made":[32],"hide":[34],"latency":[36],"and":[37,109,113,143,159],"maintain":[38],"strong":[39,110],"scalability":[40],"even":[41],"for":[42,52,65,118],"dense":[43],"matrix":[44,120],"multiplication.":[45],"Matrix":[46],"multiplication":[47,121],"an":[49,63],"ideal":[50],"candidate":[51],"benchmarking":[53],"supercomputers.":[57],"The":[58,78,150],"Cerebras":[59],"CS-2":[60],"system":[61,86],"accelerator":[64],"deep":[66],"learning":[67],"world's":[70],"largest":[71],"chip,":[72],"wafer-scale":[74],"engine":[75],"2":[76],"(WSE-2).":[77],"WSE-2":[79],"can":[80],"be":[81],"considered":[82],"a":[83,96,115,144],"distributed":[84],"memory":[85],"that":[87],"comes":[88],"745,":[90],"500":[91],"processing":[92],"elements":[93],"connected":[94],"in":[95],"low-latency":[97],"2-D":[98],"mesh":[99],"topology.":[100],"This":[101],"paper":[102],"presents":[103],"effective":[105],"maximum":[106,128],"weak":[108,145],"proposes":[114],"model":[117,158],"single-precision":[119],"on":[122],"CS-2.":[124],"We":[125],"observed":[126],"349.0TFlops/s":[131],"(matrix":[132],"size:":[133],"33,000\u00d733,000,":[134],"used":[135],"PEs:":[136],"750\u00d7750),":[137],"per":[139],"watt":[140],"79.66GFlops/W,":[142],"efficiency":[147],"1.00.":[149],"mean":[151],"absolute":[152],"percentage":[153],"error":[154],"between":[155],"our":[156],"actual":[161],"measurement":[162],"was":[163],"9.2%.":[164]},"counts_by_year":[],"updated_date":"2026-08-18T07:49:30.821534","created_date":"2026-02-15T00:00:00"}