{"id":"https://openalex.org/W2963786636","doi":"https://doi.org/10.1109/infocom.2019.8737367","title":"MG-WFBP: Efficient Data Communication for Distributed Synchronous SGD Algorithms","display_name":"MG-WFBP: Efficient Data Communication for Distributed Synchronous SGD Algorithms","publication_year":2019,"publication_date":"2019-04-01","ids":{"openalex":"https://openalex.org/W2963786636","doi":"https://doi.org/10.1109/infocom.2019.8737367","mag":"2963786636"},"language":"en","primary_location":{"id":"doi:10.1109/infocom.2019.8737367","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom.2019.8737367","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2019 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"},"type":"preprint","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/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":true,"raw_author_name":"Shaohuai Shi","raw_affiliation_strings":["Department of Computer Science, Hong Kong Baptist University"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, Hong Kong Baptist University","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":["Department of Computer Science, Hong Kong Baptist University"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, Hong Kong Baptist University","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100374506","display_name":"Bo Li","orcid":"https://orcid.org/0000-0003-2083-9105"},"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":["Department of Computer Science and Engineering, The Hong Kong University of Science and Technology"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, The Hong Kong University of Science and Technology","institution_ids":["https://openalex.org/I200769079"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5016836702"],"corresponding_institution_ids":["https://openalex.org/I141568987"],"apc_list":null,"apc_paid":null,"fwci":8.827,"has_fulltext":false,"cited_by_count":98,"citation_normalized_percentile":{"value":0.98154597,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"172","last_page":"180"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9993000030517578,"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"}},"topics":[{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9993000030517578,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9976000189781189,"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/T12676","display_name":"Machine Learning and ELM","score":0.9968000054359436,"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":0.8547186851501465},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.7191689014434814},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.698858380317688},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.6205917596817017},{"id":"https://openalex.org/keywords/node","display_name":"Node (physics)","score":0.5751352906227112},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5621281266212463},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.5193644165992737},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4873143136501312},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4286339282989502},{"id":"https://openalex.org/keywords/gpu-cluster","display_name":"GPU cluster","score":0.4181879758834839},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.35829079151153564},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3449662923812866},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.13957199454307556}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8547186851501465},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.7191689014434814},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.698858380317688},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.6205917596817017},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.5751352906227112},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5621281266212463},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.5193644165992737},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4873143136501312},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4286339282989502},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.4181879758834839},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35829079151153564},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3449662923812866},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.13957199454307556},{"id":"https://openalex.org/C66938386","wikidata":"https://www.wikidata.org/wiki/Q633538","display_name":"Structural engineering","level":1,"score":0.0},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/infocom.2019.8737367","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom.2019.8737367","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2019 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-98931","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-98931","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":null,"raw_type":"Conference paper"},{"id":"pmh:oai:repository.ust.hk:1783.1-98931","is_oa":false,"landing_page_url":"http://gateway.isiknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcAuth=LinksAMR&SrcApp=PARTNER_APP&DestLinkType=FullRecord&DestApp=WOS&KeyUT=000480426400021","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":"Conference paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":43,"referenced_works":["https://openalex.org/W1512808804","https://openalex.org/W1572016165","https://openalex.org/W1964981582","https://openalex.org/W2097117768","https://openalex.org/W2108598243","https://openalex.org/W2127941149","https://openalex.org/W2131613942","https://openalex.org/W2155893237","https://openalex.org/W2166221608","https://openalex.org/W2168231600","https://openalex.org/W2194775991","https://openalex.org/W2339765813","https://openalex.org/W2498764059","https://openalex.org/W2521757593","https://openalex.org/W2580688187","https://openalex.org/W2617766261","https://openalex.org/W2622263826","https://openalex.org/W2744387122","https://openalex.org/W2745269232","https://openalex.org/W2752512710","https://openalex.org/W2767393956","https://openalex.org/W2774000609","https://openalex.org/W2787998955","https://openalex.org/W2884711234","https://openalex.org/W2919632853","https://openalex.org/W2920397365","https://openalex.org/W2962758826","https://openalex.org/W2963339410","https://openalex.org/W2963803379","https://openalex.org/W2964004663","https://openalex.org/W2964324519","https://openalex.org/W2982475424","https://openalex.org/W4301239768","https://openalex.org/W6679154944","https://openalex.org/W6684859321","https://openalex.org/W6738460352","https://openalex.org/W6739622702","https://openalex.org/W6739693220","https://openalex.org/W6744307745","https://openalex.org/W6746839373","https://openalex.org/W6748645090","https://openalex.org/W6753209298","https://openalex.org/W6757714641"],"related_works":["https://openalex.org/W2115605526","https://openalex.org/W3093883775","https://openalex.org/W1539246760","https://openalex.org/W2786746258","https://openalex.org/W4225893763","https://openalex.org/W2788727425","https://openalex.org/W2405196115","https://openalex.org/W2104893957","https://openalex.org/W2971074373","https://openalex.org/W3159389381"],"abstract_inverted_index":{"Distributed":[0],"synchronous":[1],"stochastic":[2],"gradient":[3],"descent":[4],"has":[5],"been":[6],"widely":[7],"used":[8],"to":[9,57,85],"train":[10],"deep":[11,41,107],"neural":[12,42],"networks":[13,43],"on":[14,30,61,114,126],"computer":[15],"clusters.":[16],"With":[17],"the":[18,31,62,76,87,133],"increase":[19],"of":[20,48,55],"computational":[21],"power,":[22],"network":[23],"communications":[24],"have":[25,44],"become":[26],"one":[27,73],"limiting":[28],"factor":[29],"system":[32],"scalability.":[33],"In":[34],"this":[35],"paper,":[36],"we":[37,80],"observe":[38],"that":[39,64,132],"many":[40],"a":[45,52,71,127],"large":[46],"number":[47],"layers":[49],"with":[50,119],"only":[51],"small":[53],"amount":[54],"data":[56],"be":[58],"communicated.":[59],"Based":[60],"fact":[63],"merging":[65],"some":[66],"short":[67],"communication":[68,78],"tasks":[69],"into":[70],"single":[72],"may":[74],"reduce":[75],"overall":[77],"time,":[79],"formulate":[81],"an":[82,93,115],"optimization":[83],"problem":[84],"minimize":[86],"training":[88],"iteration":[89],"time.":[90],"We":[91],"develop":[92],"optimal":[94],"solution":[95],"named":[96],"merged-gradient":[97],"wait-free":[98],"backpropagation":[99],"(MG-WFBP)":[100],"and":[101,122,146],"implement":[102],"it":[103],"in":[104],"our":[105],"open-source":[106],"learning":[108],"platform":[109],"B-Caffe.":[110],"Our":[111],"experimental":[112],"results":[113,125],"8-node":[116],"GPU":[117],"cluster":[118,129],"10GbE":[120],"interconnect":[121],"trace-based":[123],"simulation":[124],"64-node":[128],"both":[130],"show":[131],"MG-WFBP":[134],"algorithm":[135],"can":[136],"achieve":[137],"much":[138],"better":[139],"scaling":[140],"efficiency":[141],"than":[142],"existing":[143],"methods":[144],"WFBP":[145],"SyncEASGD.":[147]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":7},{"year":2024,"cited_by_count":11},{"year":2023,"cited_by_count":17},{"year":2022,"cited_by_count":11},{"year":2021,"cited_by_count":27},{"year":2020,"cited_by_count":17},{"year":2019,"cited_by_count":6}],"updated_date":"2026-03-25T14:56:36.534964","created_date":"2025-10-10T00:00:00"}