{"id":"https://openalex.org/W4401807668","doi":"https://doi.org/10.1109/tc.2024.3449103","title":"BiRD: Bi-Directional Input Reuse Dataflow for Enhancing Depthwise Convolution Performance on Systolic Arrays","display_name":"BiRD: Bi-Directional Input Reuse Dataflow for Enhancing Depthwise Convolution Performance on Systolic Arrays","publication_year":2024,"publication_date":"2024-08-23","ids":{"openalex":"https://openalex.org/W4401807668","doi":"https://doi.org/10.1109/tc.2024.3449103"},"language":"en","primary_location":{"id":"doi:10.1109/tc.2024.3449103","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tc.2024.3449103","pdf_url":null,"source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","0016-9340","1557-9956","2326-3814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computers","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/A5017389283","display_name":"Mingeon Park","orcid":null},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Mingeon Park","raw_affiliation_strings":["Department of Computer Science and Engineering, Sungkyunkwan University, Suwon, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Sungkyunkwan University, Suwon, South Korea","institution_ids":["https://openalex.org/I848706"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113345044","display_name":"Seokjin Hwang","orcid":null},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seokjin Hwang","raw_affiliation_strings":["Department of Computer Science and Engineering, Sungkyunkwan University, Suwon, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Sungkyunkwan University, Suwon, South Korea","institution_ids":["https://openalex.org/I848706"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5061945152","display_name":"Hyungmin Cho","orcid":"https://orcid.org/0000-0001-8705-7066"},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyungmin Cho","raw_affiliation_strings":["Department of Computer Science and Engineering, Sungkyunkwan University, Suwon, South Korea"],"raw_orcid":"https://orcid.org/0000-0001-8705-7066","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Sungkyunkwan University, Suwon, South Korea","institution_ids":["https://openalex.org/I848706"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I848706"],"apc_list":{"value":2495,"currency":"USD","value_usd":2495},"apc_paid":null,"fwci":5.0006,"has_fulltext":false,"cited_by_count":16,"citation_normalized_percentile":{"value":0.96299255,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":"73","issue":"12","first_page":"2708","last_page":"2721"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.9882000088691711,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.9882000088691711,"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/T10829","display_name":"Interconnection Networks and Systems","score":0.9610000252723694,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9318000078201294,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/dataflow","display_name":"Dataflow","score":0.7518428564071655},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.7518268823623657},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7013248801231384},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.5698853731155396},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5192973017692566},{"id":"https://openalex.org/keywords/systolic-array","display_name":"Systolic array","score":0.4638322591781616},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3608747720718384},{"id":"https://openalex.org/keywords/arithmetic","display_name":"Arithmetic","score":0.3392341732978821},{"id":"https://openalex.org/keywords/very-large-scale-integration","display_name":"Very-large-scale integration","score":0.2349662482738495},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.2016185224056244},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.1927499771118164},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.14134332537651062}],"concepts":[{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.7518428564071655},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.7518268823623657},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7013248801231384},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.5698853731155396},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5192973017692566},{"id":"https://openalex.org/C150741067","wikidata":"https://www.wikidata.org/wiki/Q2377218","display_name":"Systolic array","level":3,"score":0.4638322591781616},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3608747720718384},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.3392341732978821},{"id":"https://openalex.org/C14580979","wikidata":"https://www.wikidata.org/wiki/Q876049","display_name":"Very-large-scale integration","level":2,"score":0.2349662482738495},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2016185224056244},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.1927499771118164},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.14134332537651062},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.0},{"id":"https://openalex.org/C18903297","wikidata":"https://www.wikidata.org/wiki/Q7150","display_name":"Ecology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tc.2024.3449103","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tc.2024.3449103","pdf_url":null,"source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","0016-9340","1557-9956","2326-3814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computers","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.41999998688697815,"display_name":"Climate action","id":"https://metadata.un.org/sdg/13"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W2067523571","https://openalex.org/W2531409750","https://openalex.org/W2604319603","https://openalex.org/W2606722458","https://openalex.org/W2754249189","https://openalex.org/W2887936511","https://openalex.org/W2900228909","https://openalex.org/W2911491685","https://openalex.org/W2945146780","https://openalex.org/W2963125010","https://openalex.org/W2963163009","https://openalex.org/W2963821229","https://openalex.org/W2963918968","https://openalex.org/W2964444661","https://openalex.org/W2979937837","https://openalex.org/W3016769527","https://openalex.org/W3017228913","https://openalex.org/W3017521908","https://openalex.org/W3047390932","https://openalex.org/W3097528158","https://openalex.org/W3187481008","https://openalex.org/W3190062760","https://openalex.org/W3192708755","https://openalex.org/W3200280098","https://openalex.org/W3213528054","https://openalex.org/W3217357178","https://openalex.org/W4280536985","https://openalex.org/W4280622459","https://openalex.org/W4280635517","https://openalex.org/W4280641199","https://openalex.org/W4285146087","https://openalex.org/W4288083528","https://openalex.org/W4291910419","https://openalex.org/W4297775537","https://openalex.org/W4312305797","https://openalex.org/W6737664043","https://openalex.org/W6744651773","https://openalex.org/W6758823024","https://openalex.org/W6762718338","https://openalex.org/W6796931752","https://openalex.org/W6802648153"],"related_works":["https://openalex.org/W2293118914","https://openalex.org/W2998381397","https://openalex.org/W4236419692","https://openalex.org/W3167919718","https://openalex.org/W4251718783","https://openalex.org/W2171015181","https://openalex.org/W4239447582","https://openalex.org/W2807127337","https://openalex.org/W4391183532","https://openalex.org/W2026533826"],"abstract_inverted_index":{"Depthwise":[0],"convolution":[1],"(DWConv)":[2],"is":[3,23],"an":[4],"effective":[5],"technique":[6],"for":[7,143,155],"reducing":[8],"the":[9,87,180,186],"size":[10],"and":[11,60,71,114,170],"computational":[12],"requirements":[13],"of":[14,35,68,158,182,188],"convolutional":[15],"neural":[16],"networks.":[17],"However,":[18],"DWConv's":[19],"input":[20,58,69],"reuse":[21,59,70],"pattern":[22],"not":[24],"easily":[25],"transformed":[26],"into":[27],"dense":[28],"matrix":[29,147],"multiplications,":[30],"leading":[31],"to":[32,56,76,119,146],"low":[33],"utilization":[34],"processing":[36],"elements":[37],"(PEs)":[38],"on":[39,86,125,190],"existing":[40,93],"systolic":[41,50,81,128,191],"arrays.":[42,192],"In":[43],"this":[44],"paper,":[45],"we":[46],"introduce":[47],"a":[48,77,126,133,152,162,171],"novel":[49],"array":[51],"dataflow":[52,94,121,141],"mechanism":[53],"calledBiRD,":[54],"designed":[55],"maximize":[57],"boost":[61],"DWConv":[62,156,189],"performance.":[63],"BiRD":[64,85,100,150,183],"utilizes":[65],"two":[66],"directions":[67],"necessitates":[72],"only":[73],"minor":[74],"modifications":[75],"typical":[78],"weight-stationary":[79],"type":[80],"array.":[82],"We":[83],"evaluate":[84],"Gemmini":[88],"platform,":[89],"comparing":[90],"it":[91,130],"with":[92],"types.":[95,122],"The":[96],"results":[97,131,178],"demonstrate":[98],"that":[99],"achieves":[101,151],"significant":[102],"performance":[103,187],"improvements":[104],"in":[105,132,161,165,174,184],"computation":[106,168],"time":[107,169],"reduction,":[108],"while":[109],"incurring":[110],"minimal":[111],"area":[112,135],"overhead":[113],"improved":[115],"energy":[116,175],"consumption":[117],"compared":[118],"other":[120,140],"For":[123],"example,":[124],"32$\\times{}$32":[127],"array,":[129],"9.8%":[134],"overhead,":[136],"significantly":[137],"smaller":[138],"than":[139],"types":[142],"DWConv.":[144],"Compared":[145],"multiplication-based":[148],"DWConv,":[149],"4.7$\\times{}$performance":[153],"improvement":[154],"layers":[157],"MobileNet-V2,":[159],"resulting":[160],"55.8%":[163],"reduction":[164,173],"total":[166],"inference":[167],"44.9%":[172],"consumption.":[176],"Our":[177],"highlight":[179],"effectiveness":[181],"enhancing":[185]},"counts_by_year":[{"year":2026,"cited_by_count":7},{"year":2025,"cited_by_count":9}],"updated_date":"2026-08-28T12:50:07.497085","created_date":"2025-10-10T00:00:00"}