{"id":"https://openalex.org/W3197816522","doi":"https://doi.org/10.1145/3458817.3476223","title":"Characterization and prediction of deep learning workloads in large-scale GPU datacenters","display_name":"Characterization and prediction of deep learning workloads in large-scale GPU datacenters","publication_year":2021,"publication_date":"2021-10-21","ids":{"openalex":"https://openalex.org/W3197816522","doi":"https://doi.org/10.1145/3458817.3476223","mag":"3197816522"},"language":"en","primary_location":{"id":"doi:10.1145/3458817.3476223","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3458817.3476223","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2109.01313","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5035921543","display_name":"Qinghao Hu","orcid":"https://orcid.org/0000-0003-1034-7502"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Qinghao Hu","raw_affiliation_strings":["Nanyang Technological University and Nanyang Technological University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University and Nanyang Technological University","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100614090","display_name":"Peng Sun","orcid":"https://orcid.org/0000-0002-1227-5533"},"institutions":[{"id":"https://openalex.org/I4210128910","display_name":"Group Sense (China)","ror":"https://ror.org/036wd5777","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210128910"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Peng Sun","raw_affiliation_strings":["SenseTime"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SenseTime","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049910854","display_name":"Shengen Yan","orcid":"https://orcid.org/0009-0005-3858-7972"},"institutions":[{"id":"https://openalex.org/I4210128910","display_name":"Group Sense (China)","ror":"https://ror.org/036wd5777","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210128910"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shengen Yan","raw_affiliation_strings":["SenseTime"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SenseTime","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041572550","display_name":"Yonggang Wen","orcid":"https://orcid.org/0000-0002-2751-5114"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yonggang Wen","raw_affiliation_strings":["Nanyang Technological University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101591101","display_name":"Tianwei Zhang","orcid":"https://orcid.org/0000-0001-6595-6650"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Tianwei Zhang","raw_affiliation_strings":["Nanyang Technological University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University","institution_ids":["https://openalex.org/I172675005"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":125,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"15"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.9998999834060669,"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":0.9998999834060669,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9962999820709229,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9962000250816345,"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":0.807232677936554},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.6653118133544922},{"id":"https://openalex.org/keywords/gpu-cluster","display_name":"GPU cluster","score":0.6049945950508118},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5317726731300354},{"id":"https://openalex.org/keywords/job-scheduler","display_name":"Job scheduler","score":0.5278815031051636},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.5075098872184753},{"id":"https://openalex.org/keywords/cluster","display_name":"Cluster (spacecraft)","score":0.49610671401023865},{"id":"https://openalex.org/keywords/service","display_name":"Service (business)","score":0.4481191635131836},{"id":"https://openalex.org/keywords/computer-cluster","display_name":"Computer cluster","score":0.42332857847213745},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.36095595359802246},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.20245009660720825},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.15478515625},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.11930239200592041},{"id":"https://openalex.org/keywords/operations-management","display_name":"Operations management","score":0.10047286748886108}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.807232677936554},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6653118133544922},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.6049945950508118},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5317726731300354},{"id":"https://openalex.org/C111873713","wikidata":"https://www.wikidata.org/wiki/Q1641413","display_name":"Job scheduler","level":3,"score":0.5278815031051636},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5075098872184753},{"id":"https://openalex.org/C164866538","wikidata":"https://www.wikidata.org/wiki/Q367351","display_name":"Cluster (spacecraft)","level":2,"score":0.49610671401023865},{"id":"https://openalex.org/C2780378061","wikidata":"https://www.wikidata.org/wiki/Q25351891","display_name":"Service (business)","level":2,"score":0.4481191635131836},{"id":"https://openalex.org/C29140674","wikidata":"https://www.wikidata.org/wiki/Q206637","display_name":"Computer cluster","level":2,"score":0.42332857847213745},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36095595359802246},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.20245009660720825},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.15478515625},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.11930239200592041},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.10047286748886108},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C136264566","wikidata":"https://www.wikidata.org/wiki/Q159810","display_name":"Economy","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3458817.3476223","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3458817.3476223","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2109.01313","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.01313","pdf_url":"https://arxiv.org/pdf/2109.01313","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2109.01313","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.01313","pdf_url":"https://arxiv.org/pdf/2109.01313","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":84,"referenced_works":["https://openalex.org/W1464363888","https://openalex.org/W1582774210","https://openalex.org/W1585314083","https://openalex.org/W1596936080","https://openalex.org/W1978252943","https://openalex.org/W1980489165","https://openalex.org/W1982922561","https://openalex.org/W1999984505","https://openalex.org/W2001496424","https://openalex.org/W2002472616","https://openalex.org/W2016210396","https://openalex.org/W2025549137","https://openalex.org/W2044503966","https://openalex.org/W2044853018","https://openalex.org/W2056528600","https://openalex.org/W2074084090","https://openalex.org/W2074988668","https://openalex.org/W2083842231","https://openalex.org/W2100346277","https://openalex.org/W2100830825","https://openalex.org/W2104993419","https://openalex.org/W2105947650","https://openalex.org/W2129542763","https://openalex.org/W2137726616","https://openalex.org/W2145287260","https://openalex.org/W2145398804","https://openalex.org/W2163961697","https://openalex.org/W2169729064","https://openalex.org/W2248732043","https://openalex.org/W2284351912","https://openalex.org/W2309679942","https://openalex.org/W2512971201","https://openalex.org/W2521597241","https://openalex.org/W2522820499","https://openalex.org/W2528415359","https://openalex.org/W2529165666","https://openalex.org/W2576355797","https://openalex.org/W2582319559","https://openalex.org/W2611060033","https://openalex.org/W2614112306","https://openalex.org/W2734941459","https://openalex.org/W2747599906","https://openalex.org/W2748757733","https://openalex.org/W2749299494","https://openalex.org/W2760837370","https://openalex.org/W2764100055","https://openalex.org/W2768348081","https://openalex.org/W2783425555","https://openalex.org/W2798515322","https://openalex.org/W2799240483","https://openalex.org/W2886481383","https://openalex.org/W2899071864","https://openalex.org/W2910100551","https://openalex.org/W2919594608","https://openalex.org/W2945148926","https://openalex.org/W2963403751","https://openalex.org/W2964308564","https://openalex.org/W3005780259","https://openalex.org/W3010543330","https://openalex.org/W3010637564","https://openalex.org/W3012121966","https://openalex.org/W3021027963","https://openalex.org/W3022298203","https://openalex.org/W3022548332","https://openalex.org/W3030163527","https://openalex.org/W3037110041","https://openalex.org/W3037681773","https://openalex.org/W3047783725","https://openalex.org/W3080999573","https://openalex.org/W3095922049","https://openalex.org/W3096583839","https://openalex.org/W3096956001","https://openalex.org/W3119866685","https://openalex.org/W3133465160","https://openalex.org/W3136172274","https://openalex.org/W3152476980","https://openalex.org/W3154142431","https://openalex.org/W3156353042","https://openalex.org/W3175707606","https://openalex.org/W4206374842","https://openalex.org/W4246762826","https://openalex.org/W4254234063","https://openalex.org/W4254751698","https://openalex.org/W4310895557"],"related_works":["https://openalex.org/W44553394","https://openalex.org/W4386430090","https://openalex.org/W2098509631","https://openalex.org/W1503498281","https://openalex.org/W2123689981","https://openalex.org/W2375959457","https://openalex.org/W2371048081","https://openalex.org/W2359321921","https://openalex.org/W1562505571","https://openalex.org/W2035708349"],"abstract_inverted_index":{"Modern":[0],"GPU":[1],"datacenters":[2],"are":[3],"critical":[4],"for":[5],"delivering":[6],"Deep":[7],"Learning":[8],"(DL)":[9],"models":[10],"and":[11,18,28,46,60,86],"services":[12],"in":[13],"both":[14],"the":[15,43,55,81,91,121],"research":[16],"community":[17],"industry.":[19],"When":[20],"operating":[21],"a":[22,39,51,66,98,114,132],"datacenter,":[23],"optimization":[24],"of":[25,42,57,69,83],"resource":[26,61],"scheduling":[27,116],"management":[29],"can":[30,89,119],"bring":[31],"significant":[32],"financial":[33],"benefits.":[34],"Achieving":[35],"this":[36],"goal":[37],"requires":[38],"deep":[40],"understanding":[41],"job":[44,71,124],"features":[45],"user":[47],"behaviors.":[48],"We":[49,75],"present":[50],"comprehensive":[52],"study":[53],"about":[54],"characteristics":[56],"DL":[58],"jobs":[59,85],"management.":[62],"First,":[63],"we":[64,96,111],"perform":[65],"large-scale":[67],"analysis":[68],"real-world":[70],"traces":[72],"from":[73,80],"SenseTime.":[74],"uncover":[76],"some":[77],"interesting":[78],"conclusions":[79],"perspectives":[82],"clusters,":[84],"users,":[87],"which":[88,101,118,137],"facilitate":[90],"cluster":[92,140],"system":[93],"designs.":[94],"Second,":[95],"introduce":[97],"general-purpose":[99],"framework,":[100],"manages":[102],"resources":[103],"based":[104],"on":[105],"historical":[106],"data.":[107],"As":[108],"case":[109],"studies,":[110],"design":[112],"(1)":[113],"Quasi-Shortest-Service-First":[115],"service,":[117,136],"minimize":[120],"cluster-wide":[122],"average":[123],"completion":[125],"time":[126],"by":[127,142],"up":[128,143],"to":[129,144],"6.5\u00d7;":[130],"(2)":[131],"Cluster":[133],"Energy":[134],"Saving":[135],"improves":[138],"overall":[139],"utilization":[141],"13%.":[145]},"counts_by_year":[{"year":2026,"cited_by_count":10},{"year":2025,"cited_by_count":39},{"year":2024,"cited_by_count":31},{"year":2023,"cited_by_count":28},{"year":2022,"cited_by_count":15},{"year":2021,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2021-09-13T00:00:00"}