{"id":"https://openalex.org/W3134991928","doi":"https://doi.org/10.1109/tpds.2021.3064966","title":"Elastic Deep Learning in Multi-Tenant GPU Clusters","display_name":"Elastic Deep Learning in Multi-Tenant GPU Clusters","publication_year":2021,"publication_date":"2021-03-09","ids":{"openalex":"https://openalex.org/W3134991928","doi":"https://doi.org/10.1109/tpds.2021.3064966","mag":"3134991928"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2021.3064966","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2021.3064966","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/A5113951316","display_name":"Yidi Wu","orcid":null},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yidi Wu","raw_affiliation_strings":["Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032934593","display_name":"Kaihao Ma","orcid":"https://orcid.org/0009-0001-3176-7816"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Kaihao Ma","raw_affiliation_strings":["Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100367774","display_name":"Xiao Yan","orcid":"https://orcid.org/0000-0002-2122-915X"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xiao Yan","raw_affiliation_strings":["Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0002-2122-915X","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067939728","display_name":"Liu Zhi","orcid":"https://orcid.org/0000-0001-6183-498X"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zhi Liu","raw_affiliation_strings":["Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000220537","display_name":"Zhenkun Cai","orcid":"https://orcid.org/0000-0002-0199-4866"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zhenkun Cai","raw_affiliation_strings":["Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034586871","display_name":"Yuzhen Huang","orcid":"https://orcid.org/0000-0002-9536-7918"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yuzhen Huang","raw_affiliation_strings":["Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016082884","display_name":"James Cheng","orcid":"https://orcid.org/0000-0001-6313-6288"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"James Cheng","raw_affiliation_strings":["Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011606889","display_name":"Han Yuan","orcid":"https://orcid.org/0000-0002-9636-9376"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Han Yuan","raw_affiliation_strings":["Huawei Technologies Co. Ltd, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huawei Technologies Co. Ltd, Shenzhen, China","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100616402","display_name":"Fan Yu","orcid":"https://orcid.org/0000-0002-4635-5486"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fan Yu","raw_affiliation_strings":["Huawei Technologies Co. Ltd, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huawei Technologies Co. Ltd, Shenzhen, China","institution_ids":["https://openalex.org/I2250955327"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":9.8883,"has_fulltext":false,"cited_by_count":45,"citation_normalized_percentile":{"value":0.97966238,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":"33","issue":"1","first_page":"144","last_page":"158"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.9991000294685364,"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":0.9991000294685364,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9979000091552734,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9969000220298767,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8930132389068604},{"id":"https://openalex.org/keywords/gpu-cluster","display_name":"GPU cluster","score":0.7816044092178345},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6397050619125366},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.6343216896057129},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5960775017738342},{"id":"https://openalex.org/keywords/elasticity","display_name":"Elasticity (physics)","score":0.5283311605453491},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.4665975570678711},{"id":"https://openalex.org/keywords/general-purpose-computing-on-graphics-processing-units","display_name":"General-purpose computing on graphics processing units","score":0.4518895745277405},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4479696452617645},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.4424033463001251},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4393061399459839},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.4077218472957611},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.21965250372886658},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.16170015931129456}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8930132389068604},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.7816044092178345},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6397050619125366},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.6343216896057129},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5960775017738342},{"id":"https://openalex.org/C121854251","wikidata":"https://www.wikidata.org/wiki/Q62932","display_name":"Elasticity (physics)","level":2,"score":0.5283311605453491},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4665975570678711},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.4518895745277405},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4479696452617645},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4424033463001251},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4393061399459839},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.4077218472957611},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.21965250372886658},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.16170015931129456},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2021.3064966","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2021.3064966","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":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.6299999952316284}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":86,"referenced_works":["https://openalex.org/W192446467","https://openalex.org/W1686810756","https://openalex.org/W1999955156","https://openalex.org/W2057332538","https://openalex.org/W2060393849","https://openalex.org/W2082171780","https://openalex.org/W2105947650","https://openalex.org/W2112796928","https://openalex.org/W2119391823","https://openalex.org/W2119738171","https://openalex.org/W2120422789","https://openalex.org/W2141992894","https://openalex.org/W2151373442","https://openalex.org/W2163961697","https://openalex.org/W2183341477","https://openalex.org/W2186615578","https://openalex.org/W2194775991","https://openalex.org/W2336721351","https://openalex.org/W2339765813","https://openalex.org/W2402144811","https://openalex.org/W2521140588","https://openalex.org/W2579247884","https://openalex.org/W2605643718","https://openalex.org/W2606552463","https://openalex.org/W2617242334","https://openalex.org/W2622263826","https://openalex.org/W2749988060","https://openalex.org/W2772151590","https://openalex.org/W2787998955","https://openalex.org/W2792078035","https://openalex.org/W2798515322","https://openalex.org/W2885836230","https://openalex.org/W2887862501","https://openalex.org/W2892527660","https://openalex.org/W2899071864","https://openalex.org/W2899771611","https://openalex.org/W2906007643","https://openalex.org/W2919594608","https://openalex.org/W2927905684","https://openalex.org/W2953384591","https://openalex.org/W2955146308","https://openalex.org/W2962725887","https://openalex.org/W2962758826","https://openalex.org/W2962970252","https://openalex.org/W2963403751","https://openalex.org/W2963403868","https://openalex.org/W2963655672","https://openalex.org/W2963959597","https://openalex.org/W2972874238","https://openalex.org/W2975712713","https://openalex.org/W2982612373","https://openalex.org/W3037338728","https://openalex.org/W3037519745","https://openalex.org/W3158027451","https://openalex.org/W3159109662","https://openalex.org/W4231510805","https://openalex.org/W4245826835","https://openalex.org/W4288357791","https://openalex.org/W4294583674","https://openalex.org/W4301239768","https://openalex.org/W4385245566","https://openalex.org/W6607881844","https://openalex.org/W6637373629","https://openalex.org/W6639619044","https://openalex.org/W6665801690","https://openalex.org/W6684084819","https://openalex.org/W6686509673","https://openalex.org/W6713134421","https://openalex.org/W6726983090","https://openalex.org/W6738534199","https://openalex.org/W6739622702","https://openalex.org/W6739693220","https://openalex.org/W6739901393","https://openalex.org/W6743289643","https://openalex.org/W6748645090","https://openalex.org/W6752891299","https://openalex.org/W6753801579","https://openalex.org/W6754281249","https://openalex.org/W6756009870","https://openalex.org/W6756040250","https://openalex.org/W6758283263","https://openalex.org/W6759814162","https://openalex.org/W6762211661","https://openalex.org/W6765429308","https://openalex.org/W6767869616","https://openalex.org/W6779740189"],"related_works":["https://openalex.org/W1963859303","https://openalex.org/W2364044215","https://openalex.org/W2389600408","https://openalex.org/W240129890","https://openalex.org/W3048701459","https://openalex.org/W2149078538","https://openalex.org/W2080146221","https://openalex.org/W2370314112","https://openalex.org/W1912958759","https://openalex.org/W2792081825"],"abstract_inverted_index":{"We":[0,72,121],"study":[1],"how":[2],"to":[3,10,55,106,132],"support":[4],"elasticity,":[5],"that":[6,103,125],"is,":[7],"the":[8,13,16,108,133],"ability":[9],"dynamically":[11],"adjust":[12],"parallelism":[14,111],"(i.e.,":[15],"number":[17],"of":[18,110],"GPUs),":[19],"for":[20,40],"deep":[21,78,91],"neural":[22],"network":[23],"(DNN)":[24],"training":[25],"in":[26,37,136],"a":[27,81],"GPU":[28,34,52,137],"cluster.":[29],"Elasticity":[30],"can":[31,85,127],"benefit":[32],"multi-tenant":[33],"cluster":[35,56,138],"management":[36],"many":[38],"ways,":[39],"example,":[41],"achieving":[42],"various":[43],"scheduling":[44],"objectives":[45],"(e.g.,":[46],"job":[47,49,67],"throughput,":[48],"completion":[50],"time,":[51],"efficiency)":[53],"according":[54],"load":[57],"variations,":[58],"utilizing":[59],"transient":[60],"idle":[61],"resources,":[62],"and":[63,69,84,97,117],"supporting":[64],"performance":[65],"profiling,":[66],"migration,":[68],"straggler":[70],"mitigation.":[71],"propose":[73],"EDL,":[74],"which":[75],"enables":[76],"elastic":[77],"learning":[79,92],"with":[80,89,123],"simple":[82],"API":[83],"be":[86],"easily":[87],"integrated":[88],"existing":[90],"frameworks":[93],"such":[94,113],"as":[95,114],"TensorFlow":[96],"PyTorch.":[98],"EDL":[99,126],"also":[100],"incorporates":[101],"techniques":[102],"are":[104],"necessary":[105],"reduce":[107],"overhead":[109],"adjustments,":[112],"stop-free":[115],"scaling":[116],"dynamic":[118],"data":[119],"pipeline.":[120],"demonstrate":[122],"experiments":[124],"indeed":[128],"bring":[129],"significant":[130],"benefits":[131],"above-listed":[134],"applications":[135],"management.":[139]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":12},{"year":2023,"cited_by_count":13},{"year":2022,"cited_by_count":6},{"year":2021,"cited_by_count":4}],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-10-10T00:00:00"}