{"id":"https://openalex.org/W4401567922","doi":"https://doi.org/10.23919/date58400.2024.10546719","title":"Discovering Efficient Fused Layer Configurations for Executing Multi-Workloads on Multi-Core NPUs","display_name":"Discovering Efficient Fused Layer Configurations for Executing Multi-Workloads on Multi-Core NPUs","publication_year":2024,"publication_date":"2024-03-25","ids":{"openalex":"https://openalex.org/W4401567922","doi":"https://doi.org/10.23919/date58400.2024.10546719"},"language":"en","primary_location":{"id":"doi:10.23919/date58400.2024.10546719","is_oa":false,"landing_page_url":"http://dx.doi.org/10.23919/date58400.2024.10546719","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 Design, Automation & Test in Europe Conference & Exhibition (DATE)","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/A5101990169","display_name":"Younghyun Lee","orcid":"https://orcid.org/0000-0001-8429-1324"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Younghyun Lee","raw_affiliation_strings":["Hanyang University,Dept. of Artificial Intelligence"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanyang University,Dept. of Artificial Intelligence","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111290210","display_name":"Hyejun Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I177023625","display_name":"Hansei University","ror":"https://ror.org/01bxsr356","country_code":"KR","type":"education","lineage":["https://openalex.org/I177023625"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyejun Kim","raw_affiliation_strings":["Yeonsei University,Dept. of Computer Science and Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yeonsei University,Dept. of Computer Science and Engineering","institution_ids":["https://openalex.org/I177023625"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085503041","display_name":"Yongseung Yu","orcid":"https://orcid.org/0000-0001-9838-4310"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yongseung Yu","raw_affiliation_strings":["Hanyang University,Dept. of Computer Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanyang University,Dept. of Computer Science","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018609148","display_name":"Myeongjin Cho","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Myeongjin Cho","raw_affiliation_strings":["SAPEON Korea Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SAPEON Korea Inc","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100657252","display_name":"Jiwon Seo","orcid":"https://orcid.org/0000-0002-4855-5609"},"institutions":[{"id":"https://openalex.org/I4575257","display_name":"Hanyang University","ror":"https://ror.org/046865y68","country_code":"KR","type":"education","lineage":["https://openalex.org/I4575257"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jiwon Seo","raw_affiliation_strings":["Hanyang University,Dept. of Computer Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hanyang University,Dept. of Computer Science","institution_ids":["https://openalex.org/I4575257"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086980677","display_name":"Yongjun Park","orcid":"https://orcid.org/0000-0003-3725-0380"},"institutions":[{"id":"https://openalex.org/I177023625","display_name":"Hansei University","ror":"https://ror.org/01bxsr356","country_code":"KR","type":"education","lineage":["https://openalex.org/I177023625"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yongjun Park","raw_affiliation_strings":["Yeonsei University,Dept. of Computer Science and Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yeonsei University,Dept. of Computer Science and Engineering","institution_ids":["https://openalex.org/I177023625"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9825999736785889,"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"}},"topics":[{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9825999736785889,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.9800000190734863,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9775999784469604,"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/computer-science","display_name":"Computer science","score":0.7372063994407654},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.6533176302909851},{"id":"https://openalex.org/keywords/core","display_name":"Core (optical fiber)","score":0.6499911546707153},{"id":"https://openalex.org/keywords/multi-core-processor","display_name":"Multi-core processor","score":0.5044237375259399},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4989023208618164},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.34685516357421875},{"id":"https://openalex.org/keywords/materials-science","display_name":"Materials science","score":0.09653961658477783},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.07975944876670837}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7372063994407654},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.6533176302909851},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.6499911546707153},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.5044237375259399},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4989023208618164},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.34685516357421875},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.09653961658477783},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.07975944876670837},{"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:10.23919/date58400.2024.10546719","is_oa":false,"landing_page_url":"http://dx.doi.org/10.23919/date58400.2024.10546719","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 Design, Automation & Test in Europe Conference & Exhibition (DATE)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320335489","display_name":"Institute for Information and Communications Technology Promotion","ror":"https://ror.org/01g0hqq23"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":18,"referenced_works":["https://openalex.org/W2604565756","https://openalex.org/W2606722458","https://openalex.org/W2790925711","https://openalex.org/W2804032941","https://openalex.org/W2883929540","https://openalex.org/W2911491685","https://openalex.org/W2917604311","https://openalex.org/W2936278485","https://openalex.org/W3016542674","https://openalex.org/W3017521908","https://openalex.org/W3101026687","https://openalex.org/W3174885745","https://openalex.org/W3211853974","https://openalex.org/W4249932213","https://openalex.org/W4285092328","https://openalex.org/W4297097426","https://openalex.org/W4321496239","https://openalex.org/W6735958303"],"related_works":["https://openalex.org/W2128523353","https://openalex.org/W2291648581","https://openalex.org/W1993191611","https://openalex.org/W2023938924","https://openalex.org/W2918840249","https://openalex.org/W1991859582","https://openalex.org/W2110053126","https://openalex.org/W2079303253","https://openalex.org/W2104702637","https://openalex.org/W4248099758"],"abstract_inverted_index":{"As":[0,156],"the":[1,21,73,78,99,107,140,153,160,180,185,200,210],"AI":[2,15],"industry":[3],"grows":[4],"rapidly,":[5],"Neural":[6,65],"Processing":[7],"Units":[8],"(NPUs)":[9],"have":[10],"been":[11],"developed":[12],"to":[13,30,97,116,206],"deliver":[14],"services":[16],"more":[17],"efficiently.":[18],"One":[19],"of":[20,77,101,146,159],"most":[22,63],"important":[23],"challenges":[24],"for":[25],"NPUs":[26,172],"is":[27],"task":[28],"scheduling":[29,94],"minimize":[31],"off-chip":[32,135],"memory":[33,43,75,86,110,128,136,163,190,203],"accesses,":[34,44],"which":[35,56],"may":[36],"occur":[37],"significant":[38],"performance":[39,222],"overhead.":[40],"To":[41],"reduce":[42],"multiple":[45,69,102,165],"convolution":[46,103],"layers":[47,70,80,104],"can":[48,167],"be":[49,168,225],"fused":[50,53,79],"into":[51,143],"a":[52,93,118,131,144,157,217,220],"layer":[54],"group,":[55],"offers":[57],"numerous":[58],"optimization":[59],"opportunities.":[60],"However,":[61],"in":[62,84,134,216],"Convolutional":[64],"Networks":[66],"(CNNs),":[67],"when":[68],"are":[71],"fused,":[72],"on-chip":[74,109,127,162,189,202],"utilization":[76],"gradually":[81],"decreases,":[82],"resulting":[83],"non-flat":[85],"usage.":[87,111],"In":[88],"this":[89],"paper,":[90],"we":[91],"propose":[92],"search":[95,141],"algorithm":[96,114],"optimize":[98],"fusion":[100,181],"while":[105,194],"reducing":[106],"peak":[108,126],"The":[112],"proposed":[113,186],"aims":[115],"find":[117],"schedule":[119,182],"that":[120,179],"simultaneously":[121],"optimizes":[122],"execution":[123],"time":[124],"and":[125,150,209],"usage,":[129,164],"despite":[130],"slight":[132],"increase":[133],"accesses.":[137],"It":[138],"organizes":[139],"space":[142],"graph":[145],"possible":[147],"partial":[148],"schedules":[149],"then":[151],"finds":[152],"optimal":[154],"path.":[155],"result":[158],"improved":[161],"workloads":[166,208,212],"executed":[169,214],"on":[170],"multi-core":[171,218],"with":[173],"increased":[174],"throughput.":[175],"Experimental":[176],"results":[177],"show":[178],"explored":[183],"by":[184,192,197],"method":[187],"reduced":[188],"usage":[191],"39%,":[193],"increasing":[195],"latency":[196],"13%.":[198],"When":[199],"freed":[201],"was":[204],"allocated":[205],"other":[207],"two":[211],"were":[213],"concurrently":[215],"NPU,":[219],"32%":[221],"improvement":[223],"could":[224],"achieved.":[226]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}