{"id":"https://openalex.org/W3097875570","doi":"https://doi.org/10.1145/3419111.3421307","title":"Elastic parameter server load distribution in deep learning clusters","display_name":"Elastic parameter server load distribution in deep learning clusters","publication_year":2020,"publication_date":"2020-10-12","ids":{"openalex":"https://openalex.org/W3097875570","doi":"https://doi.org/10.1145/3419111.3421307","mag":"3097875570"},"language":"en","primary_location":{"id":"doi:10.1145/3419111.3421307","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3419111.3421307","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 11th ACM Symposium on Cloud Computing","raw_type":"proceedings-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/A5062504314","display_name":"Yangrui Chen","orcid":null},"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":true,"raw_author_name":"Yangrui Chen","raw_affiliation_strings":["The University of Hong Kong"],"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013856812","display_name":"Yanghua Peng","orcid":"https://orcid.org/0000-0003-3989-4358"},"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":"Yanghua Peng","raw_affiliation_strings":["The University of Hong Kong"],"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054679083","display_name":"Yixin Bao","orcid":"https://orcid.org/0000-0002-6921-2154"},"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":"Yixin Bao","raw_affiliation_strings":["The University of Hong Kong"],"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"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"],"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028120333","display_name":"Yibo Zhu","orcid":"https://orcid.org/0000-0002-9113-2660"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yibo Zhu","raw_affiliation_strings":["ByteDance Inc"],"affiliations":[{"raw_affiliation_string":"ByteDance Inc","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5054205326","display_name":"Chuanxiong Guo","orcid":"https://orcid.org/0000-0002-0730-8468"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chuanxiong Guo","raw_affiliation_strings":["ByteDance Inc"],"affiliations":[{"raw_affiliation_string":"ByteDance Inc","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5062504314"],"corresponding_institution_ids":["https://openalex.org/I889458895"],"apc_list":null,"apc_paid":null,"fwci":1.8563,"has_fulltext":false,"cited_by_count":32,"citation_normalized_percentile":{"value":0.88085636,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"507","last_page":"521"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9994999766349792,"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":0.9994999766349792,"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/T11344","display_name":"Traffic Prediction and Management Techniques","score":0.9970999956130981,"subfield":{"id":"https://openalex.org/subfields/2215","display_name":"Building and Construction"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12676","display_name":"Machine Learning and ELM","score":0.9966999888420105,"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":0.8048429489135742},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.7805522680282593},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6294218897819519},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.5260488986968994},{"id":"https://openalex.org/keywords/testbed","display_name":"Testbed","score":0.5218700170516968},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.4829503893852234},{"id":"https://openalex.org/keywords/inefficiency","display_name":"Inefficiency","score":0.42648255825042725},{"id":"https://openalex.org/keywords/aka","display_name":"AKA","score":0.4147990942001343},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.34329208731651306},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.2610918879508972},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.1420828104019165}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8048429489135742},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.7805522680282593},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6294218897819519},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5260488986968994},{"id":"https://openalex.org/C31395832","wikidata":"https://www.wikidata.org/wiki/Q1318674","display_name":"Testbed","level":2,"score":0.5218700170516968},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.4829503893852234},{"id":"https://openalex.org/C2778869765","wikidata":"https://www.wikidata.org/wiki/Q6028363","display_name":"Inefficiency","level":2,"score":0.42648255825042725},{"id":"https://openalex.org/C121158502","wikidata":"https://www.wikidata.org/wiki/Q4652161","display_name":"AKA","level":2,"score":0.4147990942001343},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34329208731651306},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.2610918879508972},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.1420828104019165},{"id":"https://openalex.org/C161191863","wikidata":"https://www.wikidata.org/wiki/Q199655","display_name":"Library science","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/C175444787","wikidata":"https://www.wikidata.org/wiki/Q39072","display_name":"Microeconomics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3419111.3421307","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3419111.3421307","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 11th ACM Symposium on Cloud Computing","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W1994616650","https://openalex.org/W2019380637","https://openalex.org/W2057332538","https://openalex.org/W2060393849","https://openalex.org/W2097117768","https://openalex.org/W2141561793","https://openalex.org/W2143612262","https://openalex.org/W2166221608","https://openalex.org/W2194775991","https://openalex.org/W2523435939","https://openalex.org/W2579247884","https://openalex.org/W2605643718","https://openalex.org/W2609354095","https://openalex.org/W2612026221","https://openalex.org/W2798515322","https://openalex.org/W2804032941","https://openalex.org/W2911382970","https://openalex.org/W2919897868","https://openalex.org/W2920397365","https://openalex.org/W2962684017","https://openalex.org/W2964923275","https://openalex.org/W2972874238","https://openalex.org/W2975712713","https://openalex.org/W2982664135","https://openalex.org/W3047537431","https://openalex.org/W3101426332","https://openalex.org/W3102385572","https://openalex.org/W4252347102","https://openalex.org/W6713134421"],"related_works":["https://openalex.org/W2883256816","https://openalex.org/W2171408034","https://openalex.org/W3003320923","https://openalex.org/W2106140982","https://openalex.org/W2152313554","https://openalex.org/W2064303750","https://openalex.org/W4285042611","https://openalex.org/W1509300825","https://openalex.org/W3092582874","https://openalex.org/W2338718585"],"abstract_inverted_index":{"In":[0],"distributed":[1,83],"DNN":[2],"training,":[3],"parameter":[4,18,31,45,71,101,105],"servers":[5,102],"(PS)":[6],"can":[7],"become":[8],"performance":[9],"bottlenecks":[10],"due":[11],"to":[12,76,96,120,128,150],"PS":[13,63,78,88,117,140],"stragglers,":[14],"caused":[15],"by":[16],"imbalanced":[17],"distribution,":[19],"bandwidth":[20],"contention,":[21],"or":[22],"computation":[23],"interference.":[24],"Few":[25],"existing":[26],"studies":[27],"have":[28],"investigated":[29],"efficient":[30],"(aka":[32],"load)":[33],"distribution":[34,74,106],"among":[35,107],"PSs.":[36],"We":[37,66,112,132],"observe":[38],"significant":[39],"training":[40,59,85,130,155],"inefficiency":[41],"with":[42,61,125,156],"the":[43,87,110,129],"current":[44],"assignment":[46],"in":[47,86,98,153],"representative":[48],"machine":[49],"learning":[50],"frameworks":[51],"(e.g.,":[52],"MXNet,":[53],"TensorFlow),":[54],"and":[55,81,99,103,144],"big":[56],"potential":[57],"for":[58,158],"acceleration":[60],"better":[62],"load":[64,73],"distribution.":[65],"design":[67,114],"PSLD,":[68,157],"a":[69],"dynamic":[70],"server":[72],"scheme,":[75],"mitigate":[77],"straggler":[79,164],"issues":[80],"accelerate":[82],"model":[84,154],"architecture.":[89],"An":[90],"exploitation-exploration":[91],"method":[92],"is":[93],"carefully":[94],"designed":[95],"scale":[97],"out":[100,122],"adjust":[104],"PSs":[108],"on":[109,136],"go.":[111],"also":[113],"an":[115],"elastic":[116],"scaling":[118],"module":[119,135],"carry":[121],"our":[123,134],"scheme":[124],"little":[126],"interruption":[127],"process.":[131],"implement":[133],"top":[137],"of":[138],"open-source":[139],"architectures,":[141],"including":[142],"MXNet":[143],"BytePS.":[145],"Testbed":[146],"experiments":[147],"show":[148],"up":[149],"2.86x":[151],"speed-up":[152],"different":[159],"ML":[160],"models":[161],"under":[162],"various":[163],"settings.":[165]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":10},{"year":2023,"cited_by_count":10},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":5}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}