{"id":"https://openalex.org/W3036987943","doi":"https://doi.org/10.1109/tnet.2020.2999377","title":"A Scalable, High-Performance, and Fault-Tolerant Network Architecture for Distributed Machine Learning","display_name":"A Scalable, High-Performance, and Fault-Tolerant Network Architecture for Distributed Machine Learning","publication_year":2020,"publication_date":"2020-06-19","ids":{"openalex":"https://openalex.org/W3036987943","doi":"https://doi.org/10.1109/tnet.2020.2999377","mag":"3036987943"},"language":"en","primary_location":{"id":"doi:10.1109/tnet.2020.2999377","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnet.2020.2999377","pdf_url":null,"source":{"id":"https://openalex.org/S62238642","display_name":"IEEE/ACM Transactions on Networking","issn_l":"1063-6692","issn":["1063-6692","1558-2566"],"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/ACM Transactions on Networking","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/A5101517637","display_name":"Songtao Wang","orcid":"https://orcid.org/0000-0002-2073-8235"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Songtao Wang","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-2073-8235","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100380787","display_name":"Dan Li","orcid":"https://orcid.org/0000-0003-1418-7171"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dan Li","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060417049","display_name":"Cheng Yang","orcid":"https://orcid.org/0000-0001-7821-0030"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yang Cheng","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078941654","display_name":"Jinkun Geng","orcid":"https://orcid.org/0000-0002-6574-8349"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinkun Geng","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-6574-8349","affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037037671","display_name":"Yanshu Wang","orcid":"https://orcid.org/0000-0001-8473-8165"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanshu Wang","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100328245","display_name":"Shuai Wang","orcid":"https://orcid.org/0000-0001-8897-9476"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shuai Wang","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034104790","display_name":"Shu\u2010Tao Xia","orcid":"https://orcid.org/0000-0002-8639-982X"},"institutions":[{"id":"https://openalex.org/I4210114105","display_name":"Tsinghua\u2013Berkeley Shenzhen Institute","ror":"https://ror.org/02hhwwz98","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210114105","https://openalex.org/I95457486","https://openalex.org/I99065089"]},{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shutao Xia","raw_affiliation_strings":["Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0002-8639-982X","affiliations":[{"raw_affiliation_string":"Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China","institution_ids":["https://openalex.org/I4210114105","https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":null,"display_name":"Jianping Wu","orcid":null},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianping Wu","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.7327,"has_fulltext":false,"cited_by_count":13,"citation_normalized_percentile":{"value":0.67458569,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":"28","issue":"4","first_page":"1752","last_page":"1764"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9973000288009644,"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"}},"topics":[{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9973000288009644,"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"}},{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9933000206947327,"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"}},{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.9890999794006348,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.8742461204528809},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.671082079410553},{"id":"https://openalex.org/keywords/testbed","display_name":"Testbed","score":0.6660319566726685},{"id":"https://openalex.org/keywords/remote-direct-memory-access","display_name":"Remote direct memory access","score":0.6268940567970276},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.5939334034919739},{"id":"https://openalex.org/keywords/fault-tolerance","display_name":"Fault tolerance","score":0.5646185278892517},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.5543288588523865},{"id":"https://openalex.org/keywords/synchronization","display_name":"Synchronization (alternating current)","score":0.5232205390930176},{"id":"https://openalex.org/keywords/network-topology","display_name":"Network topology","score":0.4637283682823181},{"id":"https://openalex.org/keywords/load-balancing","display_name":"Load balancing (electrical power)","score":0.4563596248626709},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.43628591299057007},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.23116543889045715}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8742461204528809},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.671082079410553},{"id":"https://openalex.org/C31395832","wikidata":"https://www.wikidata.org/wiki/Q1318674","display_name":"Testbed","level":2,"score":0.6660319566726685},{"id":"https://openalex.org/C130795937","wikidata":"https://www.wikidata.org/wiki/Q2561570","display_name":"Remote direct memory access","level":2,"score":0.6268940567970276},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5939334034919739},{"id":"https://openalex.org/C63540848","wikidata":"https://www.wikidata.org/wiki/Q3140932","display_name":"Fault tolerance","level":2,"score":0.5646185278892517},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.5543288588523865},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.5232205390930176},{"id":"https://openalex.org/C199845137","wikidata":"https://www.wikidata.org/wiki/Q145490","display_name":"Network topology","level":2,"score":0.4637283682823181},{"id":"https://openalex.org/C138959212","wikidata":"https://www.wikidata.org/wiki/Q1806783","display_name":"Load balancing (electrical power)","level":3,"score":0.4563596248626709},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.43628591299057007},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.23116543889045715},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C187691185","wikidata":"https://www.wikidata.org/wiki/Q2020720","display_name":"Grid","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tnet.2020.2999377","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnet.2020.2999377","pdf_url":null,"source":{"id":"https://openalex.org/S62238642","display_name":"IEEE/ACM Transactions on Networking","issn_l":"1063-6692","issn":["1063-6692","1558-2566"],"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/ACM Transactions on Networking","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Decent work and economic growth","id":"https://metadata.un.org/sdg/8","score":0.47999998927116394}],"awards":[{"id":"https://openalex.org/G2211006180","display_name":null,"funder_award_id":"2019YFB,1802600","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"},{"id":"https://openalex.org/G5963489573","display_name":"\u4e91\u6570\u636e\u4e2d\u5fc3\u7f51\u7edc\u7684QoS\u4fdd\u969c\u65b9\u6cd5\u7814\u7a76","funder_award_id":"61672499","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6659788202","display_name":"\u6d41\u5f0f\u5927\u6570\u636e\u5904\u7406\u7684\u7f51\u7edc\u6027\u80fd\u4f18\u5316\u7814\u7a76","funder_award_id":"61772305","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/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W308556676","https://openalex.org/W1441884673","https://openalex.org/W1442374986","https://openalex.org/W1498436455","https://openalex.org/W1528469369","https://openalex.org/W1572401739","https://openalex.org/W1686810756","https://openalex.org/W1843937266","https://openalex.org/W2057332538","https://openalex.org/W2060393849","https://openalex.org/W2126210439","https://openalex.org/W2130531694","https://openalex.org/W2131613942","https://openalex.org/W2143065961","https://openalex.org/W2155861865","https://openalex.org/W2186615578","https://openalex.org/W2402144811","https://openalex.org/W2475334473","https://openalex.org/W2498764059","https://openalex.org/W2579247884","https://openalex.org/W2604783387","https://openalex.org/W2622263826","https://openalex.org/W2745734086","https://openalex.org/W2794670651","https://openalex.org/W2953384591","https://openalex.org/W2962747323","https://openalex.org/W2962758826","https://openalex.org/W2962885409","https://openalex.org/W4238465620","https://openalex.org/W4251044834","https://openalex.org/W4256524787","https://openalex.org/W6628339314","https://openalex.org/W6628377381","https://openalex.org/W6631564300","https://openalex.org/W6638593160","https://openalex.org/W6686509673","https://openalex.org/W6713134421","https://openalex.org/W6736413256","https://openalex.org/W6739622702","https://openalex.org/W6739693220","https://openalex.org/W6743955621"],"related_works":["https://openalex.org/W3090586438","https://openalex.org/W2782433361","https://openalex.org/W2416075414","https://openalex.org/W2475302168","https://openalex.org/W4366999533","https://openalex.org/W2134569538","https://openalex.org/W3198746232","https://openalex.org/W4365450377","https://openalex.org/W4244799386","https://openalex.org/W2344311498"],"abstract_inverted_index":{"In":[0,18],"large-scale":[1],"distributed":[2],"machine":[3],"learning":[4],"(DML),":[5],"the":[6,13,57,88,114],"network":[7,30,55,62],"performance":[8,89],"between":[9],"machines":[10],"significantly":[11],"impacts":[12],"speed":[14],"of":[15,34,90,98,118],"iterative":[16],"training.":[17],"this":[19],"paper":[20],"we":[21],"propose":[22],"BML,":[23],"a":[24,46,53,60,94,104],"scalable,":[25],"high-performance":[26],"and":[27,36,44,79,100],"fault-tolerant":[28],"DML":[29,119],"architecture":[31],"on":[32,41,103],"top":[33],"Ethernet":[35],"commodity":[37],"devices.":[38],"BML":[39,61,91,112],"builds":[40],"BCube":[42],"topology,":[43],"runs":[45],"fully-distributed":[47],"gradient":[48,71,130],"synchronization":[49,77,131],"algorithm.":[50,132],"Compared":[51],"to":[52,65,82,123],"Fat-Tree":[54,127],"with":[56,106,126],"same":[58],"size,":[59],"is":[63],"expected":[64],"take":[66],"much":[67],"less":[68],"time":[69,78,117],"for":[70,73],"synchronization,":[72],"both":[74],"low":[75],"theoretical":[76],"its":[80],"benefit":[81],"RDMA":[83],"transport.":[84],"With":[85],"server/link":[86],"failures,":[87],"degrades":[92],"in":[93],"graceful":[95],"way.":[96],"Experiments":[97],"MNIST":[99],"VGG-19":[101],"benchmarks":[102],"testbed":[105],"9":[107],"dual-GPU":[108],"servers":[109],"show":[110],"that,":[111],"reduces":[113],"job":[115],"completion":[116],"training":[120],"by":[121],"up":[122],"56.4%":[124],"compared":[125],"running":[128],"state-of-the-art":[129]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":2},{"year":2020,"cited_by_count":1}],"updated_date":"2026-08-01T09:00:35.917206","created_date":"2025-10-10T00:00:00"}