{"id":"https://openalex.org/W3112483307","doi":"https://doi.org/10.1145/3423211.3425675","title":"Fast Training of Deep Learning Models over Multiple GPUs","display_name":"Fast Training of Deep Learning Models over Multiple GPUs","publication_year":2020,"publication_date":"2020-12-07","ids":{"openalex":"https://openalex.org/W3112483307","doi":"https://doi.org/10.1145/3423211.3425675","mag":"3112483307"},"language":"en","primary_location":{"id":"doi:10.1145/3423211.3425675","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3423211.3425675","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st International Middleware Conference","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/A5020679643","display_name":"Xiaodong Yi","orcid":"https://orcid.org/0000-0001-7860-9543"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xiaodong Yi","raw_affiliation_strings":["The University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048036564","display_name":"Ziyue Luo","orcid":"https://orcid.org/0000-0002-1253-8137"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Ziyue Luo","raw_affiliation_strings":["The University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100633086","display_name":"Chen Meng","orcid":"https://orcid.org/0000-0002-5875-1400"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chen Meng","raw_affiliation_strings":["Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100707460","display_name":"Mengdi Wang","orcid":"https://orcid.org/0000-0002-2101-9507"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mengdi Wang","raw_affiliation_strings":["Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103437255","display_name":"Guoping Long","orcid":"https://orcid.org/0009-0006-3176-7572"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Guoping Long","raw_affiliation_strings":["Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012597518","display_name":"Chuan Wu","orcid":"https://orcid.org/0000-0002-3144-4398"},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Chuan Wu","raw_affiliation_strings":["The University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100695453","display_name":"Jun Yang","orcid":"https://orcid.org/0000-0002-3365-4989"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jun Yang","raw_affiliation_strings":["Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100636012","display_name":"Wei Lin","orcid":"https://orcid.org/0000-0002-3003-0150"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wei Lin","raw_affiliation_strings":["Alibaba Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Group","institution_ids":["https://openalex.org/I4210095624"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.4523,"has_fulltext":false,"cited_by_count":18,"citation_normalized_percentile":{"value":0.88303875,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"105","last_page":"118"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9994000196456909,"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":0.9994000196456909,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9977999925613403,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T13553","display_name":"Age of Information Optimization","score":0.9970999956130981,"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/computer-science","display_name":"Computer science","score":0.8794097900390625},{"id":"https://openalex.org/keywords/testbed","display_name":"Testbed","score":0.8313033580780029},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.6147265434265137},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5449889302253723},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5305269956588745},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.4567902684211731},{"id":"https://openalex.org/keywords/data-parallelism","display_name":"Data parallelism","score":0.4533277750015259},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.43300601840019226},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4323955178260803},{"id":"https://openalex.org/keywords/critical-path-method","display_name":"Critical path method","score":0.428458034992218},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.4146009087562561},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3956669867038727},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.36889201402664185},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.33637383580207825},{"id":"https://openalex.org/keywords/parallelism","display_name":"Parallelism (grammar)","score":0.2948891520500183},{"id":"https://openalex.org/keywords/software-engineering","display_name":"Software engineering","score":0.14512965083122253},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.09213805198669434}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8794097900390625},{"id":"https://openalex.org/C31395832","wikidata":"https://www.wikidata.org/wiki/Q1318674","display_name":"Testbed","level":2,"score":0.8313033580780029},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.6147265434265137},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5449889302253723},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5305269956588745},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4567902684211731},{"id":"https://openalex.org/C61483411","wikidata":"https://www.wikidata.org/wiki/Q3124522","display_name":"Data parallelism","level":3,"score":0.4533277750015259},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.43300601840019226},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4323955178260803},{"id":"https://openalex.org/C115874739","wikidata":"https://www.wikidata.org/wiki/Q825377","display_name":"Critical path method","level":2,"score":0.428458034992218},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.4146009087562561},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3956669867038727},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36889201402664185},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.33637383580207825},{"id":"https://openalex.org/C2781172179","wikidata":"https://www.wikidata.org/wiki/Q853109","display_name":"Parallelism (grammar)","level":2,"score":0.2948891520500183},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.14512965083122253},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.09213805198669434},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3423211.3425675","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3423211.3425675","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st International Middleware Conference","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Decent work and economic growth","id":"https://metadata.un.org/sdg/8","score":0.5899999737739563}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W1218987319","https://openalex.org/W1686810756","https://openalex.org/W2043331236","https://openalex.org/W2057392949","https://openalex.org/W2060393849","https://openalex.org/W2149294210","https://openalex.org/W2163605009","https://openalex.org/W2168231600","https://openalex.org/W2183341477","https://openalex.org/W2194775991","https://openalex.org/W2402144811","https://openalex.org/W2525778437","https://openalex.org/W2770444568","https://openalex.org/W2786066748","https://openalex.org/W2798515322","https://openalex.org/W2804682296","https://openalex.org/W2883830791","https://openalex.org/W2890720247","https://openalex.org/W2891017939","https://openalex.org/W2919594608","https://openalex.org/W2952926545","https://openalex.org/W2963351145","https://openalex.org/W2963403868","https://openalex.org/W2964308564","https://openalex.org/W2991040477","https://openalex.org/W3101104221","https://openalex.org/W6756379755"],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2164382479","https://openalex.org/W2883256816","https://openalex.org/W98480971","https://openalex.org/W2150291671","https://openalex.org/W2027972911","https://openalex.org/W2177838837","https://openalex.org/W2963831937","https://openalex.org/W2748614007","https://openalex.org/W2080419754"],"abstract_inverted_index":{"This":[0],"paper":[1],"proposes":[2],"FastT,":[3],"a":[4,16,48,107],"transparent":[5],"module":[6],"to":[7,38,57,65,70,112,128,135,143],"work":[8],"with":[9,42,155],"the":[10,40,98,114,132,146,162],"TensorFlow":[11],"framework":[12],"for":[13,30,103,120,165],"automatically":[14],"identifying":[15],"satisfying":[17],"deployment":[18],"and":[19,92,117,123,140,158],"execution":[20,93,118],"order":[21,94,119],"of":[22,74,82,109,169],"operations":[23,75,130],"in":[24,47,76,131,149],"DNN":[25,170],"models":[26,171],"over":[27],"multiple":[28],"GPUs,":[29],"expedited":[31],"model":[32,141],"training.":[33,104],"We":[34,105,152],"propose":[35],"white-box":[36],"algorithms":[37,111],"compute":[39,113],"strategies":[41,157,164],"small":[43],"computing":[44,83,100],"resource":[45],"consumption":[46],"short":[49],"time.":[50],"Recently,":[51],"similar":[52],"studies":[53],"have":[54],"been":[55],"done":[56],"optimize":[58,71],"device":[59,72,90,115],"placement":[60,73,91,116],"using":[61,79,97],"reinforcement":[62],"learning.":[63],"Compared":[64],"those":[66],"works":[67],"which":[68],"learn":[69],"several":[77],"hours":[78],"large":[80],"amounts":[81],"resources,":[84],"our":[85],"approach":[86],"can":[87],"find":[88],"excellent":[89],"within":[95],"minutes":[96],"same":[99],"node":[101],"as":[102],"design":[106,125],"list":[108],"scheduling":[110],"each":[121,150],"operation":[122],"also":[124],"an":[126],"algorithm":[127],"split":[129],"critical":[133],"path":[134],"support":[136],"fine-grained":[137],"(mixed)":[138],"data":[139],"parallelism":[142],"further":[144],"improve":[145],"training":[147,166],"speed":[148],"iteration.":[151],"compare":[153],"FastT":[154],"representative":[156],"obtain":[159],"insights":[160],"on":[161,173],"best":[163],"different":[167],"types":[168],"based":[172],"extensive":[174],"testbed":[175],"experiments.":[176]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":6},{"year":2022,"cited_by_count":9},{"year":2021,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}