{"id":"https://openalex.org/W4407404548","doi":"https://doi.org/10.1109/tcad.2025.3541486","title":"MTrain: Enable Efficient CNN Training on Heterogeneous FPGA-Based Edge Servers","display_name":"MTrain: Enable Efficient CNN Training on Heterogeneous FPGA-Based Edge Servers","publication_year":2025,"publication_date":"2025-02-12","ids":{"openalex":"https://openalex.org/W4407404548","doi":"https://doi.org/10.1109/tcad.2025.3541486"},"language":"en","primary_location":{"id":"doi:10.1109/tcad.2025.3541486","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2025.3541486","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"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 Computer-Aided Design of Integrated Circuits 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/A5028350504","display_name":"Yue Tang","orcid":"https://orcid.org/0000-0002-3253-8003"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Yue Tang","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I170201317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030875484","display_name":"Alex K. Jones","orcid":"https://orcid.org/0000-0001-7498-0206"},"institutions":[{"id":"https://openalex.org/I70983195","display_name":"Syracuse University","ror":"https://ror.org/025r5qe02","country_code":"US","type":"education","lineage":["https://openalex.org/I70983195"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alex K. Jones","raw_affiliation_strings":["Department of Electrical Engineering and the Computer Science Department, Syracuse University, Syracuse, NY, USA","Department of Electrical Engineering and Computer Science DepartmentCenter for Science and TechnologySyracuse University"],"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering and the Computer Science Department, Syracuse University, Syracuse, NY, USA","institution_ids":["https://openalex.org/I70983195"]},{"raw_affiliation_string":"Department of Electrical Engineering and Computer Science DepartmentCenter for Science and TechnologySyracuse University","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030156276","display_name":"Jinjun Xiong","orcid":"https://orcid.org/0000-0002-2620-4859"},"institutions":[{"id":"https://openalex.org/I63190737","display_name":"University at Buffalo, State University of New York","ror":"https://ror.org/01y64my43","country_code":"US","type":"education","lineage":["https://openalex.org/I63190737"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinjun Xiong","raw_affiliation_strings":["Department of Computer Science and Engineering, University at Buffalo, Buffalo, NY, USA"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University at Buffalo, Buffalo, NY, USA","institution_ids":["https://openalex.org/I63190737"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063866156","display_name":"Peipei Zhou","orcid":"https://orcid.org/0000-0002-0493-1844"},"institutions":[{"id":"https://openalex.org/I27804330","display_name":"Brown University","ror":"https://ror.org/05gq02987","country_code":"US","type":"education","lineage":["https://openalex.org/I27804330"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Peipei Zhou","raw_affiliation_strings":["School of Engineering, Brown University, Providence, RI, USA"],"affiliations":[{"raw_affiliation_string":"School of Engineering, Brown University, Providence, RI, USA","institution_ids":["https://openalex.org/I27804330"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066534595","display_name":"Jingtong Hu","orcid":"https://orcid.org/0000-0003-4029-4034"},"institutions":[{"id":"https://openalex.org/I170201317","display_name":"University of Pittsburgh","ror":"https://ror.org/01an3r305","country_code":"US","type":"education","lineage":["https://openalex.org/I170201317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jingtong Hu","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I170201317"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5028350504"],"corresponding_institution_ids":["https://openalex.org/I170201317"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.01991945,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"44","issue":"9","first_page":"3395","last_page":"3408"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9943000078201294,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9943000078201294,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9897000193595886,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9840999841690063,"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/server","display_name":"Server","score":0.7997838258743286},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.7000330090522766},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6732866764068604},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.6474713087081909},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5607093572616577},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.35623055696487427},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.34614574909210205},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3278868496417999},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.3273504972457886},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.3246912956237793},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.07878261804580688}],"concepts":[{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.7997838258743286},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.7000330090522766},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6732866764068604},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.6474713087081909},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5607093572616577},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.35623055696487427},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.34614574909210205},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3278868496417999},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3273504972457886},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.3246912956237793},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.07878261804580688},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcad.2025.3541486","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2025.3541486","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"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 Computer-Aided Design of Integrated Circuits and Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3182206366","display_name":null,"funder_award_id":"2324864","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5642188625","display_name":null,"funder_award_id":"2229873","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5877662714","display_name":null,"funder_award_id":"2329704","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G6319575343","display_name":null,"funder_award_id":"2122320","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G905957171","display_name":null,"funder_award_id":"2235364","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W2194775991","https://openalex.org/W2616014673","https://openalex.org/W2893345206","https://openalex.org/W2895195713","https://openalex.org/W3004303566","https://openalex.org/W3016430712","https://openalex.org/W3018105153","https://openalex.org/W3035430964","https://openalex.org/W3174183088","https://openalex.org/W3206758449","https://openalex.org/W4213446428","https://openalex.org/W4214834240","https://openalex.org/W4293025058","https://openalex.org/W4294891981","https://openalex.org/W4297685459","https://openalex.org/W4323338380","https://openalex.org/W4379115905","https://openalex.org/W4382119135","https://openalex.org/W4382239366","https://openalex.org/W4383749466","https://openalex.org/W4384833641","https://openalex.org/W4385731976","https://openalex.org/W4394946008","https://openalex.org/W4396753592","https://openalex.org/W4401212495","https://openalex.org/W6637373629","https://openalex.org/W6684191040","https://openalex.org/W6786965630","https://openalex.org/W6811928498","https://openalex.org/W6838539104","https://openalex.org/W6841619692","https://openalex.org/W6858229868"],"related_works":["https://openalex.org/W2092530219","https://openalex.org/W2388464034","https://openalex.org/W2533125852","https://openalex.org/W2140460949","https://openalex.org/W230091440","https://openalex.org/W2105580438","https://openalex.org/W2111241003","https://openalex.org/W1967938402","https://openalex.org/W2386041993","https://openalex.org/W1608572506"],"abstract_inverted_index":{"FPGA-based":[0,17,249],"edge":[1,88,113,250],"servers":[2,21,38,251],"are":[3,39,70,222],"used":[4],"in":[5,8,228],"many":[6],"applications":[7],"smart":[9],"cities,":[10],"hospitals,":[11],"retail,":[12],"etc.":[13,36],"Equipped":[14],"with":[15,25,171,252],"heterogeneous":[16,122,248],"accelerator":[18],"cards,":[19],"the":[20,45,49,52,64,67,73,77,87,105,112,133,138,142,209,274],"can":[22,242],"be":[23,83,108],"implemented":[24],"multiple":[26],"tasks,":[27],"including":[28],"efficient":[29,244],"video":[30],"prepossessing,":[31],"machine":[32,79],"learning":[33,80],"algorithm":[34],"acceleration,":[35],"These":[37],"required":[40],"to":[41,54,56,72,86,107,130,273],"implement":[42],"inference":[43],"during":[44,51],"daytime":[46],"while":[47,181],"retraining":[48],"model":[50],"night":[53],"adapt":[55],"new":[57,61],"environments,":[58],"domains,":[59],"or":[60],"users.":[62],"During":[63],"retraining,":[65],"conventionally,":[66],"incoming":[68],"data":[69,135,157],"transmitted":[71],"cloud,":[74],"and":[75,95,141,163,185,191,233],"then":[76],"updated":[78],"models":[81,106],"will":[82],"transferred":[84],"back":[85],"server.":[89],"Such":[90],"a":[91,201,213],"process":[92,140,211],"is":[93,102,125,176],"inefficient":[94],"cannot":[96,169],"protect":[97],"users\u2019":[98],"privacy,":[99],"so":[100],"it":[101,128],"desirable":[103],"for":[104,156,230],"directly":[109],"trained":[110],"on":[111,121,224,247],"servers.":[114],"Deploying":[115],"convolutional":[116],"neural":[117],"network":[118],"(CNN)":[119],"training":[120,139,150,204,210,226,246],"resource-constrained":[123],"FPGAs":[124],"challenging":[126],"since":[127],"needs":[129],"consider":[131],"both":[132],"complex":[134],"dependency":[136],"of":[137,219],"communication":[143,193,235],"bottleneck":[144],"among":[145],"different":[146,220,225],"FPGAs.":[147],"Previous":[148],"multiaccelerator":[149,203],"algorithms":[151],"select":[152],"optimal":[153],"scheduling":[154,205],"strategies":[155],"parallelism":[158,161,165,229],"(DP),":[159],"tensor":[160],"(TP),":[162],"pipeline":[164],"(PP).":[166],"However,":[167],"PP":[168],"deal":[170],"batch":[172],"normalization":[173],"(BN)":[174],"which":[175],"an":[177],"essential":[178],"CNN":[179,245],"operator,":[180],"purely":[182],"applying":[183],"DP":[184],"TP":[186],"suffers":[187],"from":[188],"resource":[189],"under-utilization":[190],"intensive":[192],"costs.":[194],"In":[195],"this":[196],"work,":[197],"we":[198,241],"propose":[199],"MTrain,":[200],"novel":[202],"strategy":[206],"that":[207,240],"transfers":[208],"into":[212],"multibranch":[214],"workflow,":[215],"thus":[216],"independent":[217],"suboperations":[218],"branches":[221],"executed":[223],"accelerators":[227],"better":[231],"utilization":[232],"reduced":[234],"overhead.":[236],"Experimental":[237],"results":[238],"show":[239],"achieve":[243],"<inline-formula":[253],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[254,261],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">":[255,262],"<tex-math":[256,263],"notation=\"LaTeX\">$1.07\\times":[257],"$":[258,265],"</tex-math></inline-formula>":[259,266],"\u2013<inline-formula":[260],"notation=\"LaTeX\">$2.21\\times":[264],"speedup":[267],"under":[268],"15-GB/s":[269],"peer-to-peer":[270],"bandwidth":[271],"compared":[272],"state-of-the-art":[275],"work.":[276]},"counts_by_year":[],"updated_date":"2025-12-28T23:10:05.387466","created_date":"2025-10-10T00:00:00"}