{"id":"https://openalex.org/W4393178455","doi":"https://doi.org/10.1109/asp-dac58780.2024.10473976","title":"TransFRU: Efficient Deployment of Transformers on FPGA with Full Resource Utilization","display_name":"TransFRU: Efficient Deployment of Transformers on FPGA with Full Resource Utilization","publication_year":2024,"publication_date":"2024-01-22","ids":{"openalex":"https://openalex.org/W4393178455","doi":"https://doi.org/10.1109/asp-dac58780.2024.10473976"},"language":"en","primary_location":{"id":"doi:10.1109/asp-dac58780.2024.10473976","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asp-dac58780.2024.10473976","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 29th Asia and South Pacific Design Automation Conference (ASP-DAC)","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/A5001477434","display_name":"Hongji Wang","orcid":"https://orcid.org/0000-0003-3246-8583"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongji Wang","raw_affiliation_strings":["Fudan University,State Key Lab of ASIC &#x0026; System,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of ASIC &#x0026; System,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100580160","display_name":"Yueyin Bai","orcid":"https://orcid.org/0009-0006-2938-5153"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yueyin Bai","raw_affiliation_strings":["Fudan University,State Key Lab of ASIC &#x0026; System,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of ASIC &#x0026; System,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100614702","display_name":"Jun Yu","orcid":"https://orcid.org/0000-0001-5008-2153"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Yu","raw_affiliation_strings":["Fudan University,State Key Lab of ASIC &#x0026; System,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of ASIC &#x0026; System,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5107970380","display_name":"Kun Wang","orcid":"https://orcid.org/0000-0002-0750-5357"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kun Wang","raw_affiliation_strings":["Fudan University,State Key Lab of ASIC &#x0026; System,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of ASIC &#x0026; System,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I24943067"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"521","last_page":"526"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11522","display_name":"VLSI and FPGA Design Techniques","score":0.960099995136261,"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/T11522","display_name":"VLSI and FPGA Design Techniques","score":0.960099995136261,"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/T10363","display_name":"Low-power high-performance VLSI design","score":0.9595999717712402,"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/T11032","display_name":"VLSI and Analog Circuit Testing","score":0.9341999888420105,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.7406697869300842},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.6895767450332642},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.609555721282959},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5056328773498535},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.44257816672325134},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1983444094657898},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.1898353397846222},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.15806326270103455},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.082940012216568}],"concepts":[{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.7406697869300842},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.6895767450332642},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.609555721282959},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5056328773498535},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.44257816672325134},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1983444094657898},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.1898353397846222},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.15806326270103455},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.082940012216568}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asp-dac58780.2024.10473976","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asp-dac58780.2024.10473976","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 29th Asia and South Pacific Design Automation Conference (ASP-DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":18,"referenced_works":["https://openalex.org/W2613938013","https://openalex.org/W2762374354","https://openalex.org/W2949320316","https://openalex.org/W3017024317","https://openalex.org/W3043675702","https://openalex.org/W3047848469","https://openalex.org/W3130240120","https://openalex.org/W3155487259","https://openalex.org/W3159727696","https://openalex.org/W3206453033","https://openalex.org/W3206837665","https://openalex.org/W4280582299","https://openalex.org/W4308479898","https://openalex.org/W4321637080","https://openalex.org/W4383749520","https://openalex.org/W4383749557","https://openalex.org/W4384947491","https://openalex.org/W4385245566"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W2770234245","https://openalex.org/W96612179","https://openalex.org/W4229499248","https://openalex.org/W2566006169","https://openalex.org/W2987774938","https://openalex.org/W2096844293","https://openalex.org/W2363944576","https://openalex.org/W2351041855","https://openalex.org/W2570254841"],"abstract_inverted_index":{"Transformer-based":[0],"models":[1,23],"have":[2],"achieved":[3],"huge":[4],"success":[5],"in":[6,111],"various":[7],"artificial":[8],"intelligence":[9],"(AI)":[10],"tasks,":[11],"e.g.,":[12],"natural":[13],"language":[14],"processing":[15,91],"(NLP)":[16],"and":[17,58,89,108,149,160,179],"computer":[18],"vision":[19],"(CV).":[20],"However,":[21],"transformer-based":[22,63],"always":[24],"suffer":[25],"from":[26],"high":[27],"computation":[28,56,133],"density,":[29],"making":[30],"them":[31],"hard":[32],"to":[33,52,94,130],"be":[34],"deployed":[35],"on":[36],"resource-constrained":[37],"devices":[38],"like":[39],"field-programmable":[40],"gate":[41],"array":[42],"(FPGA).":[43],"Among":[44],"the":[45,55,60,106,112,132],"overall":[46],"process":[47],"of":[48,54,62,81,115,135],"transformers,":[49],"self-attention":[50,76,116],"contributes":[51],"most":[53],"load":[57],"becomes":[59],"bottleneck":[61],"models.":[64],"In":[65],"this":[66],"paper,":[67],"we":[68,85,104,169],"propose":[69],"TransFRU,":[70],"a":[71,118],"novel":[72],"FPGA-based":[73],"accelerator":[74],"for":[75,127],"mechanism":[77],"with":[78,155,166,185],"full":[79],"utilization":[80],"hardware":[82],"resources.":[83],"Specifically,":[84],"first":[86],"leverage":[87],"4-bit":[88],"8-bit":[90],"elements":[92],"(PEs)":[93],"package":[95],"multiple":[96],"signed":[97],"multiplications":[98],"into":[99],"one":[100,136],"DSP":[101,137,177],"block.":[102,138],"Second,":[103],"skip":[105],"zero":[107],"near-zero":[109],"values":[110],"intermediate":[113],"result":[114],"by":[117],"sorting":[119,122],"engine.":[120],"The":[121],"engine":[123],"is":[124],"also":[125],"responsible":[126],"operand":[128],"sharing":[129],"boost":[131],"efficiency":[134,153,164,183],"Experimental":[139],"results":[140],"show":[141],"that":[142],"our":[143],"TransFRU":[144],"achieves":[145],"$7.86-49.16":[146],"\\times$":[147,151,158,162,173,181],"speedup":[148,159],"$151.1":[150],"energy":[152,163,182],"compared":[154,165,184],"CPU,":[156],"$1.41":[157],"$5.9":[161],"GPU.":[167],"Furthermore,":[168],"observe":[170],"$1.91-":[171],"13.56":[172],"better":[174],"throughput":[175],"per":[176],"block":[178],"$3.53-9.62":[180],"previous":[186],"FPGA":[187],"accelerators.":[188]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":6}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}
