{"id":"https://openalex.org/W3107997909","doi":"https://doi.org/10.1109/mnet.011.2000530","title":"A Quantitative Survey of Communication Optimizations in Distributed Deep Learning","display_name":"A Quantitative Survey of Communication Optimizations in Distributed Deep Learning","publication_year":2020,"publication_date":"2020-12-02","ids":{"openalex":"https://openalex.org/W3107997909","doi":"https://doi.org/10.1109/mnet.011.2000530","mag":"3107997909"},"language":"en","primary_location":{"id":"doi:10.1109/mnet.011.2000530","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mnet.011.2000530","pdf_url":null,"source":{"id":"https://openalex.org/S186584794","display_name":"IEEE Network","issn_l":"0890-8044","issn":["0890-8044","1558-156X"],"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 Network","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/A5016836702","display_name":"Shaohuai Shi","orcid":"https://orcid.org/0000-0002-1418-5160"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":true,"raw_author_name":"Shaohuai Shi","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048313055","display_name":"Zhenheng Tang","orcid":"https://orcid.org/0000-0001-8769-9974"},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zhenheng Tang","raw_affiliation_strings":["Hong Kong Baptist University, Hong Kong"],"affiliations":[{"raw_affiliation_string":"Hong Kong Baptist University, Hong Kong","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100730785","display_name":"Xiaowen Chu","orcid":"https://orcid.org/0000-0001-9745-4372"},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xiaowen Chu","raw_affiliation_strings":["Hong Kong Baptist University, Hong Kong"],"affiliations":[{"raw_affiliation_string":"Hong Kong Baptist University, Hong Kong","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027626597","display_name":"Chengjian Liu","orcid":"https://orcid.org/0000-0002-0103-4670"},"institutions":[{"id":"https://openalex.org/I4210152380","display_name":"Shenzhen Technology University","ror":"https://ror.org/04qzpec27","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210152380"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chengjian Liu","raw_affiliation_strings":["Shenzhen Technology University, China"],"affiliations":[{"raw_affiliation_string":"Shenzhen Technology University, China","institution_ids":["https://openalex.org/I4210152380"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100391896","display_name":"Wei Wang","orcid":"https://orcid.org/0000-0002-1717-5785"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Wei Wang","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5060171629","display_name":"Bo Li","orcid":"https://orcid.org/0000-0003-2955-750X"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Bo Li","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5016836702"],"corresponding_institution_ids":["https://openalex.org/I200769079"],"apc_list":null,"apc_paid":null,"fwci":2.8452,"has_fulltext":false,"cited_by_count":47,"citation_normalized_percentile":{"value":0.92494764,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":"35","issue":"3","first_page":"230","last_page":"237"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9997000098228455,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9993000030517578,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8559533357620239},{"id":"https://openalex.org/keywords/infiniband","display_name":"InfiniBand","score":0.7811319231987},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5161901116371155},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.48461899161338806},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.47831061482429504},{"id":"https://openalex.org/keywords/distributed-learning","display_name":"Distributed learning","score":0.4488992691040039},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.44562211632728577},{"id":"https://openalex.org/keywords/lossless-compression","display_name":"Lossless compression","score":0.4337031841278076},{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.41424405574798584},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4004058837890625},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.34945687651634216},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.23531031608581543}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8559533357620239},{"id":"https://openalex.org/C2781030343","wikidata":"https://www.wikidata.org/wiki/Q922437","display_name":"InfiniBand","level":2,"score":0.7811319231987},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5161901116371155},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.48461899161338806},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.47831061482429504},{"id":"https://openalex.org/C2779582901","wikidata":"https://www.wikidata.org/wiki/Q21013010","display_name":"Distributed learning","level":2,"score":0.4488992691040039},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.44562211632728577},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.4337031841278076},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.41424405574798584},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4004058837890625},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.34945687651634216},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.23531031608581543},{"id":"https://openalex.org/C19417346","wikidata":"https://www.wikidata.org/wiki/Q7922","display_name":"Pedagogy","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/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.0},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"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":3,"locations":[{"id":"doi:10.1109/mnet.011.2000530","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mnet.011.2000530","pdf_url":null,"source":{"id":"https://openalex.org/S186584794","display_name":"IEEE Network","issn_l":"0890-8044","issn":["0890-8044","1558-156X"],"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 Network","raw_type":"journal-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-109725","is_oa":false,"landing_page_url":"http://www.scopus.com/record/display.url?eid=2-s2.0-85097409727&origin=inward","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Article"},{"id":"pmh:oai:repository.ust.hk:1783.1-109725","is_oa":false,"landing_page_url":"http://repository.ust.hk/ir/Record/1783.1-109725","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.6399999856948853,"display_name":"Industry, innovation and infrastructure","id":"https://metadata.un.org/sdg/9"}],"awards":[{"id":"https://openalex.org/G2855474093","display_name":null,"funder_award_id":"12200418,16206417,16207818","funder_id":"https://openalex.org/F4320321592","funder_display_name":"Research Grants Council, University Grants Committee"},{"id":"https://openalex.org/G7621131871","display_name":null,"funder_award_id":"62002240","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320321592","display_name":"Research Grants Council, University Grants Committee","ror":"https://ror.org/00djwmt25"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W2138243089","https://openalex.org/W2160054705","https://openalex.org/W2168231600","https://openalex.org/W2888206291","https://openalex.org/W2919632853","https://openalex.org/W2920397365","https://openalex.org/W2926543120","https://openalex.org/W2962758826","https://openalex.org/W2963664311","https://openalex.org/W2963786636","https://openalex.org/W2975712713","https://openalex.org/W2981114289","https://openalex.org/W2982475424","https://openalex.org/W2982664135","https://openalex.org/W3081331745","https://openalex.org/W4288357791","https://openalex.org/W4297685247","https://openalex.org/W6680402377","https://openalex.org/W6684859321","https://openalex.org/W6739693220","https://openalex.org/W6748019269","https://openalex.org/W6753958361","https://openalex.org/W6762211661"],"related_works":["https://openalex.org/W2406331183","https://openalex.org/W2363988059","https://openalex.org/W2053215237","https://openalex.org/W2792231649","https://openalex.org/W1939358748","https://openalex.org/W1995426833","https://openalex.org/W3150273738","https://openalex.org/W2613555492","https://openalex.org/W2141107779","https://openalex.org/W2350246940"],"abstract_inverted_index":{"Nowadays,":[0],"large":[1],"and":[2,53,68,79,112,129,133],"complex":[3],"deep":[4],"learning":[5,63],"(DL)":[6],"models":[7,104],"are":[8,114],"increasingly":[9],"trained":[10],"in":[11,19],"a":[12,34,81,92,137],"distributed":[13,44,88],"manner":[14],"across":[15],"multiple":[16],"worker":[17],"machines,":[18],"which":[20],"extensive":[21],"communications":[22],"between":[23],"workers":[24],"pose":[25],"serious":[26],"scaling":[27,142],"problems.":[28],"In":[29],"this":[30],"article,":[31],"we":[32],"present":[33,73],"quantitative":[35],"survey":[36],"of":[37,84,150],"communication":[38,51,76],"optimization":[39,77],"techniques":[40,78],"for":[41,153],"data":[42],"parallel":[43],"DL.":[45],"We":[46,72,99,144],"first":[47],"identify":[48],"the":[49,55,62,65,69,74,102,121,130,141,146],"major":[50],"challenges":[52],"classify":[54],"existing":[56],"solutions":[57],"into":[58],"three":[59],"levels,":[60],"namely":[61],"algorithm,":[64],"system":[66,131],"architecture,":[67],"network":[70],"infrastructure.":[71],"state-of-the-art":[75],"conduct":[80],"comparative":[82],"study":[83],"seven":[85],"common":[86],"lossless":[87,124],"DL":[89,103],"methods":[90],"on":[91,140],"32-GPU":[93],"cluster":[94],"with":[95,105,120,148],"100Gb/s":[96,127],"InfiniBand":[97],"(IB).":[98],"show":[100],"that":[101],"low":[106],"model":[107],"intensity":[108],"(such":[109],"as":[110],"BERT":[111],"BERT-Large)":[113],"difficult":[115],"to":[116],"scale":[117],"out":[118],"even":[119],"best":[122],"available":[123],"algorithm":[125],"over":[126],"IB;":[128],"architecture":[132],"scheduling":[134],"algorithms":[135],"have":[136],"critical":[138],"impact":[139],"property.":[143],"conclude":[145],"article":[147],"discussions":[149],"open":[151],"issues":[152],"further":[154],"investigation.":[155]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":9},{"year":2024,"cited_by_count":6},{"year":2023,"cited_by_count":14},{"year":2022,"cited_by_count":8},{"year":2021,"cited_by_count":7},{"year":2019,"cited_by_count":1}],"updated_date":"2026-03-28T08:17:26.163206","created_date":"2025-10-10T00:00:00"}