{"id":"https://openalex.org/W3121689374","doi":"https://doi.org/10.1109/tpds.2021.3052895","title":"DL2: A Deep Learning-Driven Scheduler for Deep Learning Clusters","display_name":"DL2: A Deep Learning-Driven Scheduler for Deep Learning Clusters","publication_year":2021,"publication_date":"2021-01-20","ids":{"openalex":"https://openalex.org/W3121689374","doi":"https://doi.org/10.1109/tpds.2021.3052895","mag":"3121689374"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2021.3052895","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2021.3052895","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Transactions on Parallel and Distributed Systems","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/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":["University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0003-3989-4358","affiliations":[{"raw_affiliation_string":"University of Hong Kong, Hong Kong, China","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":["University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-6921-2154","affiliations":[{"raw_affiliation_string":"University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001041727","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":false,"raw_author_name":"Yangrui Chen","raw_affiliation_strings":["University of Hong Kong, Hong Kong, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Hong Kong, Hong Kong, China","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":["University of Hong Kong, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-3144-4398","affiliations":[{"raw_affiliation_string":"University of Hong Kong, Hong Kong, China","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100633086","display_name":"Chen Meng","orcid":"https://orcid.org/0000-0002-5875-1400"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen Meng","raw_affiliation_strings":["NAOC, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NAOC, Beijing, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100636012","display_name":"Wei Lin","orcid":"https://orcid.org/0000-0002-3003-0150"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Lin","raw_affiliation_strings":["Alibaba Inc., Hanzhou, Zhejiang, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alibaba Inc., Hanzhou, Zhejiang, China","institution_ids":["https://openalex.org/I45928872"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":17.0698,"has_fulltext":false,"cited_by_count":85,"citation_normalized_percentile":{"value":0.99168019,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":100},"biblio":{"volume":"32","issue":"8","first_page":"1947","last_page":"1960"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.9987999796867371,"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":0.9987999796867371,"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":0.9983999729156494,"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":0.9958000183105469,"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":0.8437561988830566},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7285923957824707},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.6599034070968628},{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.6417449712753296},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.6372567415237427},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5975003242492676},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5462004542350769},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5195515751838684},{"id":"https://openalex.org/keywords/job-scheduler","display_name":"Job scheduler","score":0.5024228096008301},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.48211669921875},{"id":"https://openalex.org/keywords/job-shop-scheduling","display_name":"Job shop scheduling","score":0.41211211681365967},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.18047919869422913},{"id":"https://openalex.org/keywords/schedule","display_name":"Schedule","score":0.1358737349510193},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.11376953125},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.11199951171875}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8437561988830566},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7285923957824707},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.6599034070968628},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.6417449712753296},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6372567415237427},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5975003242492676},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5462004542350769},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5195515751838684},{"id":"https://openalex.org/C111873713","wikidata":"https://www.wikidata.org/wiki/Q1641413","display_name":"Job scheduler","level":3,"score":0.5024228096008301},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.48211669921875},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.41211211681365967},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.18047919869422913},{"id":"https://openalex.org/C68387754","wikidata":"https://www.wikidata.org/wiki/Q7271585","display_name":"Schedule","level":2,"score":0.1358737349510193},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.11376953125},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.11199951171875},{"id":"https://openalex.org/C160403385","wikidata":"https://www.wikidata.org/wiki/Q220543","display_name":"Queue","level":2,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2021.3052895","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2021.3052895","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Transactions on Parallel and Distributed Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.5899999737739563,"display_name":"Decent work and economic growth","id":"https://metadata.un.org/sdg/8"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":85,"referenced_works":["https://openalex.org/W1503189082","https://openalex.org/W1522301498","https://openalex.org/W1665214252","https://openalex.org/W1686810756","https://openalex.org/W1890643295","https://openalex.org/W2060393849","https://openalex.org/W2102646063","https://openalex.org/W2105947650","https://openalex.org/W2116873850","https://openalex.org/W2119391823","https://openalex.org/W2119717200","https://openalex.org/W2121863487","https://openalex.org/W2131876387","https://openalex.org/W2133564696","https://openalex.org/W2133569115","https://openalex.org/W2141992894","https://openalex.org/W2143612262","https://openalex.org/W2145339207","https://openalex.org/W2155027007","https://openalex.org/W2163605009","https://openalex.org/W2163961697","https://openalex.org/W2182361439","https://openalex.org/W2186615578","https://openalex.org/W2194775991","https://openalex.org/W2402144811","https://openalex.org/W2521757593","https://openalex.org/W2546571074","https://openalex.org/W2579247884","https://openalex.org/W2589642470","https://openalex.org/W2605643718","https://openalex.org/W2609354095","https://openalex.org/W2613904329","https://openalex.org/W2617411258","https://openalex.org/W2622263826","https://openalex.org/W2744628735","https://openalex.org/W2749807327","https://openalex.org/W2783239537","https://openalex.org/W2786414509","https://openalex.org/W2798515322","https://openalex.org/W2824027552","https://openalex.org/W2885836230","https://openalex.org/W2896457183","https://openalex.org/W2899071864","https://openalex.org/W2919594608","https://openalex.org/W2919897868","https://openalex.org/W2953384591","https://openalex.org/W2962684017","https://openalex.org/W2962725887","https://openalex.org/W2962835968","https://openalex.org/W2963341956","https://openalex.org/W2963549123","https://openalex.org/W2964043796","https://openalex.org/W2964121744","https://openalex.org/W2964265128","https://openalex.org/W2964308564","https://openalex.org/W2968986602","https://openalex.org/W2975712713","https://openalex.org/W3037338728","https://openalex.org/W3102385572","https://openalex.org/W4214717370","https://openalex.org/W4288357791","https://openalex.org/W6631190155","https://openalex.org/W6637242042","https://openalex.org/W6637373629","https://openalex.org/W6639249596","https://openalex.org/W6665801690","https://openalex.org/W6679434410","https://openalex.org/W6683204974","https://openalex.org/W6684084819","https://openalex.org/W6684191040","https://openalex.org/W6685961532","https://openalex.org/W6686509673","https://openalex.org/W6692846177","https://openalex.org/W6713134421","https://openalex.org/W6737778391","https://openalex.org/W6738144653","https://openalex.org/W6739622702","https://openalex.org/W6743756900","https://openalex.org/W6748687944","https://openalex.org/W6753801579","https://openalex.org/W6755207826","https://openalex.org/W6756009870","https://openalex.org/W6759814162","https://openalex.org/W6762211661","https://openalex.org/W7010963004"],"related_works":["https://openalex.org/W986318368","https://openalex.org/W2092966558","https://openalex.org/W1956651153","https://openalex.org/W1984267569","https://openalex.org/W2347561926","https://openalex.org/W2327547880","https://openalex.org/W2315766899","https://openalex.org/W2002944882","https://openalex.org/W3153517983","https://openalex.org/W210339300"],"abstract_inverted_index":{"Efficient":[0],"resource":[1,148,162],"scheduling":[2,28],"is":[3,101,122],"essential":[4],"for":[5,73,145],"maximal":[6],"utilization":[7],"of":[8,34,139,193],"expensive":[9],"deep":[10],"learning":[11,22,93,96,107,132],"(DL)":[12],"clusters.":[13],"Existing":[14],"cluster":[15,116],"schedulers":[16],"either":[17],"are":[18,41],"agnostic":[19],"to":[20,59,86],"machine":[21],"(ML)":[23],"workload":[24],"characteristics,":[25],"or":[26,44],"use":[27],"heuristics":[29],"based":[30,108],"on":[31,109,157,167],"operators'":[32],"understanding":[33],"particular":[35],"ML":[36],"framework":[37],"and":[38,63,94,143,159,182],"workload,":[39],"which":[40],"less":[42],"efficient":[43,64],"not":[45],"general":[46],"enough.":[47],"In":[48],"this":[49],"article,":[50],"we":[51,67],"show":[52],"that":[53,172],"DL":[54,74,127,141,165],"techniques":[55],"can":[56],"be":[57],"adopted":[58],"design":[60],"a":[61,70,90,98],"generic":[62],"scheduler.":[65],"Specifically,":[66],"propose":[68],"DL2,":[69],"DL-driven":[71],"scheduler":[72,176,185],"clusters,":[75],"targeting":[76],"global":[77],"training":[78,137],"job":[79,110,147,195],"expedition":[80],"by":[81,113,130,179,188],"dynamically":[82],"resizing":[83],"resources":[84],"allocated":[85],"jobs.":[87],"DL2":[88,156,173],"advocates":[89],"joint":[91],"supervised":[92,106],"reinforcement":[95,131],"approach:":[97],"neural":[99,120],"network":[100,121],"warmed":[102],"up":[103],"via":[104],"offline":[105],"traces":[111],"produced":[112],"the":[114,119,125,136,140],"existing":[115],"scheduler;":[117],"then":[118],"plugged":[123],"into":[124],"live":[126],"cluster,":[128],"fine-tuned":[129],"carried":[133],"out":[134],"throughout":[135],"progress":[138],"jobs,":[142],"used":[144],"deciding":[146],"allocation":[149],"in":[150,164,191],"an":[151],"online":[152],"fashion.":[153],"We":[154],"implement":[155],"Kubernetes":[158],"enable":[160],"dynamic":[161],"scaling":[163],"jobs":[166],"MXNet.":[168],"Extensive":[169],"evaluation":[170],"shows":[171],"outperforms":[174],"fairness":[175],"(i.e.,":[177,186],"DRF)":[178],"44.1":[180],"percent":[181,190],"expert":[183],"heuristic":[184],"Optimus)":[187],"17.5":[189],"terms":[192],"average":[194],"completion":[196],"time.":[197]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":21},{"year":2024,"cited_by_count":21},{"year":2023,"cited_by_count":23},{"year":2022,"cited_by_count":10},{"year":2021,"cited_by_count":6}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}