{"id":"https://openalex.org/W4410995768","doi":"https://doi.org/10.1145/3742437","title":"MCoreOPU: An FPGA-based Multi-Core Overlay Processor for Transformer-based Models","display_name":"MCoreOPU: An FPGA-based Multi-Core Overlay Processor for Transformer-based Models","publication_year":2025,"publication_date":"2025-06-03","ids":{"openalex":"https://openalex.org/W4410995768","doi":"https://doi.org/10.1145/3742437"},"language":"en","primary_location":{"id":"doi:10.1145/3742437","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3742437","pdf_url":null,"source":{"id":"https://openalex.org/S112809824","display_name":"ACM Transactions on Reconfigurable Technology and Systems","issn_l":"1936-7406","issn":["1936-7406","1936-7414"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Reconfigurable Technology and Systems","raw_type":"journal-article"},"type":"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/A5101262539","display_name":"Shaoqiang Lu","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I4394709157","display_name":"Eastern Institute of Technology, Ningbo","ror":"https://ror.org/036mbz113","country_code":null,"type":"education","lineage":["https://openalex.org/I4394709157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shaoqiang Lu","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China and Eastern Institute of Technology, Ningbo, China"],"raw_orcid":"https://orcid.org/0009-0002-8624-8666","affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China and Eastern Institute of Technology, Ningbo, China","institution_ids":["https://openalex.org/I183067930","https://openalex.org/I4394709157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073130132","display_name":"Tiandong Zhao","orcid":null},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tiandong Zhao","raw_affiliation_strings":["University of California, Los Angeles, California, USA"],"raw_orcid":"https://orcid.org/0009-0000-5348-1715","affiliations":[{"raw_affiliation_string":"University of California, Los Angeles, California, USA","institution_ids":["https://openalex.org/I161318765"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101612957","display_name":"Ting-Jung Lin","orcid":"https://orcid.org/0000-0002-5208-482X"},"institutions":[{"id":"https://openalex.org/I4210165339","display_name":"Ningbo Institute of Industrial Technology","ror":"https://ror.org/05nqg3g04","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165339"]},{"id":"https://openalex.org/I4394709157","display_name":"Eastern Institute of Technology, Ningbo","ror":"https://ror.org/036mbz113","country_code":null,"type":"education","lineage":["https://openalex.org/I4394709157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ting-Jung Lin","raw_affiliation_strings":["Ningbo Institute of Digital Twin, Ningbo, China and Eastern\u00a0Institute of Technology, Ningbo, China","Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, China"],"raw_orcid":"https://orcid.org/0000-0002-5208-482X","affiliations":[{"raw_affiliation_string":"Ningbo Institute of Digital Twin, Ningbo, China and Eastern\u00a0Institute of Technology, Ningbo, China","institution_ids":["https://openalex.org/I4210165339"]},{"raw_affiliation_string":"Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, China","institution_ids":["https://openalex.org/I4394709157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101712184","display_name":"Rumin Zhang","orcid":"https://orcid.org/0000-0003-2617-2380"},"institutions":[{"id":"https://openalex.org/I4210165339","display_name":"Ningbo Institute of Industrial Technology","ror":"https://ror.org/05nqg3g04","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165339"]},{"id":"https://openalex.org/I4394709157","display_name":"Eastern Institute of Technology, Ningbo","ror":"https://ror.org/036mbz113","country_code":null,"type":"education","lineage":["https://openalex.org/I4394709157"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rumin Zhang","raw_affiliation_strings":["Ningbo Institute of Digital Twin, Ningbo, China and Eastern\u00a0Institute of Technology, Ningbo, China","Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, China"],"raw_orcid":"https://orcid.org/0000-0003-2617-2380","affiliations":[{"raw_affiliation_string":"Ningbo Institute of Digital Twin, Ningbo, China and Eastern\u00a0Institute of Technology, Ningbo, China","institution_ids":["https://openalex.org/I4210165339"]},{"raw_affiliation_string":"Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, China","institution_ids":["https://openalex.org/I4394709157"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101491215","display_name":"C.-L. Wu","orcid":"https://orcid.org/0000-0002-9939-8922"},"institutions":[{"id":"https://openalex.org/I13591777","display_name":"University of Nottingham Ningbo China","ror":"https://ror.org/03y4dt428","country_code":"CN","type":"education","lineage":["https://openalex.org/I13591777","https://openalex.org/I142263535"]},{"id":"https://openalex.org/I4210165339","display_name":"Ningbo Institute of Industrial Technology","ror":"https://ror.org/05nqg3g04","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165339"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen Wu","raw_affiliation_strings":["Chiplet CAD and Manufacturing Engineering Research Center of Zhejiang Province, Ningbo, China, Ningbo Institute of Digital Twin, Ningbo, China and Eastern Institute of Technology, Ningbo, China","Engineering Research Center of Chiplet Design and Manufacturing of Zhejiang Province, Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, China"],"raw_orcid":"https://orcid.org/0000-0002-9939-8922","affiliations":[{"raw_affiliation_string":"Chiplet CAD and Manufacturing Engineering Research Center of Zhejiang Province, Ningbo, China, Ningbo Institute of Digital Twin, Ningbo, China and Eastern Institute of Technology, Ningbo, China","institution_ids":["https://openalex.org/I4210165339","https://openalex.org/I13591777"]},{"raw_affiliation_string":"Engineering Research Center of Chiplet Design and Manufacturing of Zhejiang Province, Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5008695429","display_name":"Lei He","orcid":"https://orcid.org/0000-0002-5266-3805"},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lei He","raw_affiliation_strings":["University of California, Los Angeles, California, USA"],"raw_orcid":"https://orcid.org/0000-0002-5266-3805","affiliations":[{"raw_affiliation_string":"University of California, Los Angeles, California, USA","institution_ids":["https://openalex.org/I161318765"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.5495,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.8325636,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":"18","issue":"3","first_page":"1","last_page":"27"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11338","display_name":"Advancements in Photolithography Techniques","score":0.9842000007629395,"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"}},"topics":[{"id":"https://openalex.org/T11338","display_name":"Advancements in Photolithography Techniques","score":0.9842000007629395,"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"}},{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9789999723434448,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T11522","display_name":"VLSI and FPGA Design Techniques","score":0.9768000245094299,"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/computer-science","display_name":"Computer science","score":0.8716989755630493},{"id":"https://openalex.org/keywords/overlay","display_name":"Overlay","score":0.7152732610702515},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.6568912863731384},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5184497237205505},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.5160602927207947},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.48065459728240967},{"id":"https://openalex.org/keywords/core","display_name":"Core (optical fiber)","score":0.45819056034088135},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3845010995864868},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.22044336795806885},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.08450552821159363},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.06471514701843262}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8716989755630493},{"id":"https://openalex.org/C136085584","wikidata":"https://www.wikidata.org/wiki/Q910289","display_name":"Overlay","level":2,"score":0.7152732610702515},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.6568912863731384},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5184497237205505},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.5160602927207947},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.48065459728240967},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.45819056034088135},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3845010995864868},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.22044336795806885},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.08450552821159363},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.06471514701843262},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3742437","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3742437","pdf_url":null,"source":{"id":"https://openalex.org/S112809824","display_name":"ACM Transactions on Reconfigurable Technology and Systems","issn_l":"1936-7406","issn":["1936-7406","1936-7414"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Reconfigurable Technology and Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W2896457183","https://openalex.org/W2979455536","https://openalex.org/W3008072732","https://openalex.org/W3034100311","https://openalex.org/W3047848469","https://openalex.org/W3129831491","https://openalex.org/W3177265267","https://openalex.org/W3196923642","https://openalex.org/W3207622241","https://openalex.org/W4241251112","https://openalex.org/W4300865759","https://openalex.org/W4308479898","https://openalex.org/W4380874786","https://openalex.org/W4380881077","https://openalex.org/W4385187240","https://openalex.org/W4385567276","https://openalex.org/W4390309921","https://openalex.org/W4393578753","https://openalex.org/W4393949386","https://openalex.org/W4402465042"],"related_works":["https://openalex.org/W595346907","https://openalex.org/W153296825","https://openalex.org/W598989511","https://openalex.org/W2375779923","https://openalex.org/W2041986468","https://openalex.org/W1967800214","https://openalex.org/W2055675609","https://openalex.org/W1967938402","https://openalex.org/W2386041993","https://openalex.org/W1608572506"],"abstract_inverted_index":{"Transformer-based":[0],"models":[1],"have":[2,19],"achieved":[3],"extensive":[4],"success":[5],"with":[6,163,217],"increasingly":[7],"large":[8],"numbers":[9],"of":[10,63,73,124,197],"parameters":[11],"and":[12,54,97,110,120,142,149,166,203],"computations,":[13],"for":[14,107,199],"which":[15],"many":[16],"multi-core":[17,45,139],"accelerators":[18],"been":[20],"developed.":[21],"Nevertheless,":[22],"they":[23],"suffer":[24],"from":[25],"limited":[26],"throughput":[27,194,222],"due":[28],"to":[29,50,69,76,87,116,128,144],"either":[30],"low":[31],"operating":[32,61],"frequency":[33,62],"or":[34],"high":[35],"communication":[36],"overhead":[37],"between":[38,101],"cores.":[39],"This":[40],"article":[41],"proposes":[42],"an":[43],"FPGA-based":[44,214],"overlay":[46],"processor,":[47],"named":[48],"MCoreOPU,":[49],"optimize":[51,137],"intra-core":[52,79,152],"computation":[53,153],"inter-core":[55,147],"communication.":[56],"First,":[57],"we":[58,82,113,136],"boost":[59],"the":[60,64,71,74,78,94,125,130,138,146,151,170,175,183,193,218,221],"processing":[65],"element":[66],"(PE)":[67],"array":[68,177],"double":[70],"rest":[72,184],"processor":[75],"improve":[77],"throughput.":[80],"Second,":[81],"develop":[83,121],"on-chip":[84],"synchronization":[85,115,118],"routers":[86],"reduce":[88,117,129],"off-chip":[89,131],"memory":[90,132],"traffic,":[91],"where":[92,174],"only":[93],"partial":[95],"sum":[96],"maximum":[98],"are":[99],"communicated":[100],"cores":[102,165],"rather":[103],"than":[104,212],"entire":[105],"vectors":[106],"layer":[108],"normalization":[109],"softmax.":[111],"Moreover,":[112],"pipeline":[114],"latency":[119],"a":[122],"bypass":[123],"interconnect":[126],"bus":[127],"access":[133],"latency.":[134],"Finally,":[135],"model":[140],"allocation":[141],"scheduling":[143],"minimize":[145],"communications":[148],"maximize":[150],"efficiency.":[154],"The":[155],"MCoreOPU":[156,198],"is":[157,206,227],"implemented":[158],"in":[159],"8-bit":[160],"fixed-point":[161],"precision":[162],"four":[164,167],"DDRs":[168],"on":[169],"Xilinx":[171],"U200":[172],"FPGA,":[173],"PE":[176],"runs":[178,185],"at":[179,186],"600":[180],"MHz":[181],"while":[182],"300":[187],"MHz.":[188],"Experimental":[189],"results":[190],"show":[191],"that":[192],"per":[195,223],"MAC":[196,225],"BERT,":[200],"ViT,":[201],"GPT-2,":[202],"LLaMA":[204],"inference":[205],"1.31":[207],"\\(\\times\\)":[208,210,231,233],"\u20137.18":[209],"higher":[211],"other":[213],"accelerators.":[215],"Compared":[216],"A100":[219],"GPU,":[220],"equivalent":[224],"efficiency":[226],"improved":[228],"by":[229],"22.52":[230],"\u201327.12":[232],".":[234]},"counts_by_year":[{"year":2026,"cited_by_count":3}],"updated_date":"2026-06-19T17:40:00.097472","created_date":"2025-10-10T00:00:00"}
