{"id":"https://openalex.org/W2919897868","doi":"https://doi.org/10.1109/infocom.2019.8737460","title":"Deep Learning-based Job Placement in Distributed Machine Learning Clusters","display_name":"Deep Learning-based Job Placement in Distributed Machine Learning Clusters","publication_year":2019,"publication_date":"2019-04-01","ids":{"openalex":"https://openalex.org/W2919897868","doi":"https://doi.org/10.1109/infocom.2019.8737460","mag":"2919897868"},"language":"en","primary_location":{"id":"doi:10.1109/infocom.2019.8737460","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom.2019.8737460","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":"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/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":true,"raw_author_name":"Yixin Bao","raw_affiliation_strings":["Department of Computer Science, The University of Hong Kong"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, 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":["Department of Computer Science, The University of Hong Kong"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"last","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":["Department of Computer Science, The University of Hong Kong"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5054679083"],"corresponding_institution_ids":["https://openalex.org/I889458895"],"apc_list":null,"apc_paid":null,"fwci":23.955,"has_fulltext":false,"cited_by_count":141,"citation_normalized_percentile":{"value":0.99450122,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"505","last_page":"513"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.9995999932289124,"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"}},"topics":[{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.9995999932289124,"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"}},{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9984999895095825,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9901000261306763,"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":0.8419373631477356},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6538335084915161},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6492066383361816},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5635482668876648},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.5365578532218933},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.5252465605735779},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5077158212661743},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.5064283609390259},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.18640798330307007},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.13103675842285156}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8419373631477356},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6538335084915161},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6492066383361816},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5635482668876648},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5365578532218933},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.5252465605735779},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5077158212661743},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.5064283609390259},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.18640798330307007},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.13103675842285156}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/infocom.2019.8737460","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom.2019.8737460","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"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Decent work and economic growth","id":"https://metadata.un.org/sdg/8","score":0.6800000071525574}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":55,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W1686810756","https://openalex.org/W1832693441","https://openalex.org/W2021667890","https://openalex.org/W2060393849","https://openalex.org/W2083842231","https://openalex.org/W2085797995","https://openalex.org/W2102709380","https://openalex.org/W2105947650","https://openalex.org/W2121863487","https://openalex.org/W2129542763","https://openalex.org/W2133569115","https://openalex.org/W2139258336","https://openalex.org/W2141992894","https://openalex.org/W2155027007","https://openalex.org/W2163605009","https://openalex.org/W2163961697","https://openalex.org/W2186615578","https://openalex.org/W2189062930","https://openalex.org/W2194775991","https://openalex.org/W2402144811","https://openalex.org/W2461411193","https://openalex.org/W2546571074","https://openalex.org/W2597068831","https://openalex.org/W2609354095","https://openalex.org/W2617411258","https://openalex.org/W2744628735","https://openalex.org/W2749807327","https://openalex.org/W2766447205","https://openalex.org/W2798515322","https://openalex.org/W2953384591","https://openalex.org/W2962684017","https://openalex.org/W2962725887","https://openalex.org/W2962835968","https://openalex.org/W2963549123","https://openalex.org/W2964043796","https://openalex.org/W2964121744","https://openalex.org/W3102385572","https://openalex.org/W4214717370","https://openalex.org/W4235985402","https://openalex.org/W4236491543","https://openalex.org/W6631190155","https://openalex.org/W6637373629","https://openalex.org/W6665801690","https://openalex.org/W6680929869","https://openalex.org/W6683204974","https://openalex.org/W6684084819","https://openalex.org/W6684191040","https://openalex.org/W6686509673","https://openalex.org/W6687460195","https://openalex.org/W6692846177","https://openalex.org/W6713134421","https://openalex.org/W6738144653","https://openalex.org/W6743756900","https://openalex.org/W7010963004"],"related_works":["https://openalex.org/W986318368","https://openalex.org/W1982914007","https://openalex.org/W3212136884","https://openalex.org/W2159583675","https://openalex.org/W1852113167","https://openalex.org/W1824242903","https://openalex.org/W2103851758","https://openalex.org/W2155470929","https://openalex.org/W1493858311","https://openalex.org/W2111125783"],"abstract_inverted_index":{"Production":[0],"machine":[1,16],"learning":[2,113],"(ML)":[3],"clusters":[4],"commonly":[5],"host":[6],"a":[7,74,81,92,108,144,182],"variety":[8],"of":[9,143,147,185,199],"distributed":[10],"ML":[11,29,84,179],"workloads,":[12],"e.g.,":[13],"speech":[14],"recognition,":[15],"translation.":[17],"While":[18],"server":[19],"sharing":[20],"among":[21,27],"jobs":[22,30,90],"improves":[23],"resource":[24,50],"utilization,":[25],"interference":[26,69,96],"co-located":[28],"can":[31],"lead":[32],"to":[33,125,151],"significant":[34],"performance":[35,99],"downgrade.":[36],"Existing":[37],"cluster":[38,85,184],"schedulers":[39,194],"(e.g.,":[40],"Mesos)":[41],"are":[42],"interference-oblivious":[43],"in":[44,58,91,181,197],"their":[45],"job":[46,53,201],"placement,":[47],"causing":[48],"suboptimal":[49],"efficiency.":[51],"Interference-aware":[52],"placement":[54,153],"has":[55],"been":[56],"studied":[57],"the":[59],"literature,":[60],"but":[61],"was":[62],"treated":[63],"using":[64,165,177],"detailed":[65],"workload":[66],"profiling":[67],"and":[68,97,128,138,168],"modeling,":[70],"which":[71,162],"is":[72,105,163],"not":[73],"general":[75],"solution.":[76],"This":[77],"paper":[78],"presents":[79],"Harmony,":[80],"deep":[82,111],"learning-driven":[83],"scheduler":[86],"that":[87,94,190],"places":[88],"training":[89,101,127],"manner":[93],"minimizes":[95],"maximizes":[98],"(i.e.,":[100],"completion":[102,202],"time).":[103],"Harmony":[104,191],"based":[106],"on":[107],"carefully":[109],"designed":[110],"reinforcement":[112],"(DRL)":[114],"framework":[115],"augmented":[116],"with":[117],"reward":[118,148,159,172],"modeling.":[119],"The":[120],"DRL":[121],"employs":[122],"state-of-the-art":[123],"techniques":[124],"stabilize":[126],"improve":[129],"convergence,":[130],"including":[131],"actor-critic":[132],"algorithm,":[133],"job-aware":[134],"action":[135],"space":[136],"exploration":[137],"experience":[139],"replay.":[140],"In":[141],"view":[142],"common":[145],"lack":[146],"samples":[149,167],"corresponding":[150],"different":[152],"decisions,":[154],"we":[155],"build":[156],"an":[157],"auxiliary":[158],"prediction":[160],"model,":[161],"trained":[164],"historical":[166],"used":[169],"for":[170,173],"producing":[171],"unseen":[174],"placement.":[175],"Experiments":[176],"real":[178],"workloads":[180],"Kubernetes":[183],"6":[186],"GPU":[187],"servers":[188],"show":[189],"outperforms":[192],"representative":[193],"by":[195],"25%":[196],"terms":[198],"average":[200],"time.":[203]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":21},{"year":2024,"cited_by_count":23},{"year":2023,"cited_by_count":28},{"year":2022,"cited_by_count":20},{"year":2021,"cited_by_count":24},{"year":2020,"cited_by_count":20},{"year":2019,"cited_by_count":4}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}
