{"id":"https://openalex.org/W4322706674","doi":"https://doi.org/10.1109/tpds.2023.3250462","title":"Accelerating Distributed DNN Training via Transport Layer Scheduling","display_name":"Accelerating Distributed DNN Training via Transport Layer Scheduling","publication_year":2023,"publication_date":"2023-02-28","ids":{"openalex":"https://openalex.org/W4322706674","doi":"https://doi.org/10.1109/tpds.2023.3250462"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2023.3250462","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2023.3250462","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Parallel and Distributed 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/A5036041576","display_name":"Qingyang Duan","orcid":"https://orcid.org/0000-0002-9268-1979"},"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":"Qingyang Duan","raw_affiliation_strings":["Department of Electronic Engineering, Fudan University, Shanghai, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering, Fudan University, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101468920","display_name":"Chao Peng","orcid":"https://orcid.org/0000-0002-4900-5932"},"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":"Chao Peng","raw_affiliation_strings":["Department of Electronic Engineering, Fudan University, Shanghai, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering, Fudan University, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067078448","display_name":"Zeqin Wang","orcid":null},"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":"Zeqin Wang","raw_affiliation_strings":["Department of Electronic Engineering, Fudan University, Shanghai, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering, Fudan University, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063792608","display_name":"Yuedong Xu","orcid":"https://orcid.org/0000-0003-4168-3998"},"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":"Yuedong Xu","raw_affiliation_strings":["Department of Electronic Engineering, Fudan University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0003-4168-3998","affiliations":[{"raw_affiliation_string":"Department of Electronic Engineering, Fudan University, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083783069","display_name":"Shaoteng Liu","orcid":"https://orcid.org/0000-0001-5407-2905"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shaoteng Liu","raw_affiliation_strings":["2012 Lab, Huawei Technologies Co. Ltd., Shenzhen, Guangdong, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"2012 Lab, Huawei Technologies Co. Ltd., Shenzhen, Guangdong, China","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075806425","display_name":"Jun Wu","orcid":"https://orcid.org/0000-0001-7090-8653"},"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 Wu","raw_affiliation_strings":["School of Computer Science and Technology, Fudan University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0001-7090-8653","affiliations":[{"raw_affiliation_string":"School of Computer Science and Technology, Fudan University, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5068489266","display_name":"John C. S. Lui","orcid":"https://orcid.org/0000-0001-7466-0384"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"John C. S. Lui","raw_affiliation_strings":["Computer Science &#x0026; Engineering, The Chinese University of Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0001-7466-0384","affiliations":[{"raw_affiliation_string":"Computer Science &#x0026; Engineering, The Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.9572,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.69128329,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":"34","issue":"5","first_page":"1650","last_page":"1666"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12303","display_name":"Tensor decomposition and applications","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12303","display_name":"Tensor decomposition and applications","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9904000163078308,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9887999892234802,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8404484391212463},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.7167627811431885},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.6803675889968872},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.6583666801452637},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.5511042475700378},{"id":"https://openalex.org/keywords/network-packet","display_name":"Network packet","score":0.49189019203186035},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.47532251477241516},{"id":"https://openalex.org/keywords/transport-layer","display_name":"Transport layer","score":0.4712657928466797},{"id":"https://openalex.org/keywords/mercury","display_name":"Mercury (programming language)","score":0.4603670835494995},{"id":"https://openalex.org/keywords/fair-share-scheduling","display_name":"Fair-share scheduling","score":0.41589680314064026},{"id":"https://openalex.org/keywords/partition","display_name":"Partition (number theory)","score":0.41444283723831177},{"id":"https://openalex.org/keywords/dynamic-priority-scheduling","display_name":"Dynamic priority scheduling","score":0.410590797662735},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.17669159173965454},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1724950671195984},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.12629491090774536},{"id":"https://openalex.org/keywords/quality-of-service","display_name":"Quality of service","score":0.08318302035331726}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8404484391212463},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.7167627811431885},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.6803675889968872},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6583666801452637},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5511042475700378},{"id":"https://openalex.org/C158379750","wikidata":"https://www.wikidata.org/wiki/Q214111","display_name":"Network packet","level":2,"score":0.49189019203186035},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.47532251477241516},{"id":"https://openalex.org/C7345512","wikidata":"https://www.wikidata.org/wiki/Q209372","display_name":"Transport layer","level":3,"score":0.4712657928466797},{"id":"https://openalex.org/C2777777548","wikidata":"https://www.wikidata.org/wiki/Q206040","display_name":"Mercury (programming language)","level":2,"score":0.4603670835494995},{"id":"https://openalex.org/C31689143","wikidata":"https://www.wikidata.org/wiki/Q733809","display_name":"Fair-share scheduling","level":3,"score":0.41589680314064026},{"id":"https://openalex.org/C42812","wikidata":"https://www.wikidata.org/wiki/Q1082910","display_name":"Partition (number theory)","level":2,"score":0.41444283723831177},{"id":"https://openalex.org/C107568181","wikidata":"https://www.wikidata.org/wiki/Q5319000","display_name":"Dynamic priority scheduling","level":3,"score":0.410590797662735},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.17669159173965454},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1724950671195984},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.12629491090774536},{"id":"https://openalex.org/C5119721","wikidata":"https://www.wikidata.org/wiki/Q220501","display_name":"Quality of service","level":2,"score":0.08318302035331726},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2023.3250462","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2023.3250462","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Parallel and Distributed Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/9","score":0.550000011920929,"display_name":"Industry, innovation and infrastructure"}],"awards":[{"id":"https://openalex.org/G4898904745","display_name":null,"funder_award_id":"22ZR1407000","funder_id":"https://openalex.org/F4320309612","funder_display_name":"Natural Science Foundation of Shanghai"},{"id":"https://openalex.org/G4995821630","display_name":"\u5178\u578b\u7f51\u7edc\u7cfb\u7edf\u4e2d\u6df1\u5ea6\u5f3a\u5316\u5b66\u4e60\u7684\u53ef\u89e3\u91ca\u6027\u548c\u9c81\u68d2\u6027\u7814\u7a76","funder_award_id":"62072117","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320309612","display_name":"Natural Science Foundation of Shanghai","ror":null},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":52,"referenced_works":["https://openalex.org/W2045271686","https://openalex.org/W2083842231","https://openalex.org/W2138243089","https://openalex.org/W2186615578","https://openalex.org/W2769986458","https://openalex.org/W2920397365","https://openalex.org/W2963601856","https://openalex.org/W2963786636","https://openalex.org/W2975712713","https://openalex.org/W2982475424","https://openalex.org/W2989289980","https://openalex.org/W3015640161","https://openalex.org/W3044837714","https://openalex.org/W3047002910","https://openalex.org/W3047357290","https://openalex.org/W3048581864","https://openalex.org/W3081388954","https://openalex.org/W3106914464","https://openalex.org/W3107997909","https://openalex.org/W3134069806","https://openalex.org/W3173464048","https://openalex.org/W3177554119","https://openalex.org/W3193250980","https://openalex.org/W4283207077","https://openalex.org/W4283217358","https://openalex.org/W4283219828","https://openalex.org/W4297775537","https://openalex.org/W4301239768","https://openalex.org/W6679393576","https://openalex.org/W6684191040","https://openalex.org/W6686509673","https://openalex.org/W6687566353","https://openalex.org/W6713134421","https://openalex.org/W6737664043","https://openalex.org/W6738460352","https://openalex.org/W6739693220","https://openalex.org/W6743499376","https://openalex.org/W6746081298","https://openalex.org/W6746200960","https://openalex.org/W6748645090","https://openalex.org/W6754799804","https://openalex.org/W6754818105","https://openalex.org/W6755207826","https://openalex.org/W6760703128","https://openalex.org/W6762211661","https://openalex.org/W6764309476","https://openalex.org/W6766978945","https://openalex.org/W6784425352","https://openalex.org/W6784871562","https://openalex.org/W6786422851","https://openalex.org/W6794216301","https://openalex.org/W7067822191"],"related_works":["https://openalex.org/W2383090288","https://openalex.org/W2387846155","https://openalex.org/W2184166483","https://openalex.org/W2397293317","https://openalex.org/W2377713709","https://openalex.org/W2373973180","https://openalex.org/W2372008037","https://openalex.org/W2978148977","https://openalex.org/W3003077338","https://openalex.org/W2225350526"],"abstract_inverted_index":{"Communication":[0],"scheduling":[1,42,59,95],"is":[2,25,66,73],"crucial":[3],"to":[4,38,96,137,168,183,186,194],"accelerate":[5],"the":[6,15,40,70,93,97,101,107,111,120,130,134,139,179,187],"training":[7,180],"of":[8,18,142],"large":[9],"deep":[10,20],"learning":[11],"models,":[12],"in":[13,77,110,150,161],"which":[14],"transmission":[16],"order":[17],"layer-wise":[19],"neural":[21],"network":[22,71],"(DNN)":[23],"tensors":[24],"determined":[26],"for":[27,63],"a":[28,47,86],"better":[29],"computation-communication":[30],"overlap.":[31],"Prior":[32],"approaches":[33],"adopt":[34],"user-level":[35],"tensor":[36,54,64,196],"partitioning":[37,65,197],"enhance":[39],"priority":[41,94,109],"with":[43,106],"finer":[44],"granularity.":[45,103],"However,":[46],"startup":[48],"time":[49],"slot":[50],"inserted":[51],"before":[52],"every":[53],"partition":[55],"will":[56,114],"neutralize":[57],"this":[58,81],"gain.":[60],"Tuning":[61],"hyper-parameters":[62],"difficult,":[67],"especially":[68],"when":[69],"bandwidth":[72],"shared":[74],"or":[75],"time-varying":[76],"multi-tenant":[78],"clusters.":[79],"In":[80],"article,":[82],"we":[83],"propose":[84],"Mercury,":[85],"simple":[87],"transport":[88,98,135],"layer":[89,99,136],"scheduler":[90],"that":[91,176],"moves":[92],"at":[100,133],"packet":[102],"The":[104],"packets":[105],"highest":[108],"Mercury":[112,118,149,164,177],"buffer":[113],"be":[115],"transmitted":[116],"first.":[117],"achieves":[119],"near-optimal":[121],"overlapping":[122,141],"between":[123],"communication":[124,170],"and":[125,145,152,171,191],"computation.":[126],"It":[127],"also":[128],"leverages":[129],"immediate":[131],"aggregation":[132],"enable":[138],"full":[140],"gradient":[143],"push":[144],"pull.":[146],"We":[147],"implement":[148],"MXNet":[151],"conduct":[153],"comprehensive":[154],"experiments":[155],"on":[156],"five":[157],"popular":[158],"DNN":[159],"models":[160],"various":[162],"environments.":[163],"can":[165],"well":[166],"adapt":[167],"dynamic":[169],"computation":[172],"resources.":[173],"Experiments":[174],"show":[175],"accelerates":[178],"by":[181],"up":[182],"130%":[184],"compared":[185,193],"classical":[188],"PS":[189],"architecture,":[190],"104%":[192],"state-of-the-art":[195],"methods.":[198]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}
