{"id":"https://openalex.org/W4402896827","doi":"https://doi.org/10.1109/iwqos61813.2024.10682877","title":"Scheduling Deep Learning Jobs in Multi-Tenant GPU Clusters via Wise Resource Sharing","display_name":"Scheduling Deep Learning Jobs in Multi-Tenant GPU Clusters via Wise Resource Sharing","publication_year":2024,"publication_date":"2024-06-19","ids":{"openalex":"https://openalex.org/W4402896827","doi":"https://doi.org/10.1109/iwqos61813.2024.10682877"},"language":"en","primary_location":{"id":"doi:10.1109/iwqos61813.2024.10682877","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iwqos61813.2024.10682877","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/ACM 32nd International Symposium on Quality of Service (IWQoS)","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/A5101188976","display_name":"Yizhou Luo","orcid":null},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yizhou Luo","raw_affiliation_strings":["Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)"],"affiliations":[{"raw_affiliation_string":"Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018867983","display_name":"Qiang Wang","orcid":"https://orcid.org/0000-0002-2986-967X"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qiang Wang","raw_affiliation_strings":["Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)"],"affiliations":[{"raw_affiliation_string":"Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016836702","display_name":"Shaohuai Shi","orcid":"https://orcid.org/0000-0002-1418-5160"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shaohuai Shi","raw_affiliation_strings":["Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)"],"affiliations":[{"raw_affiliation_string":"Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111342958","display_name":"Jiaxin Lai","orcid":null},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiaxin Lai","raw_affiliation_strings":["Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)"],"affiliations":[{"raw_affiliation_string":"Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050911130","display_name":"Shuhan Qi","orcid":"https://orcid.org/0000-0002-6903-145X"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shuhan Qi","raw_affiliation_strings":["Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)"],"affiliations":[{"raw_affiliation_string":"Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100417650","display_name":"Jiajia Zhang","orcid":"https://orcid.org/0000-0003-4623-6134"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiajia Zhang","raw_affiliation_strings":["Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)"],"affiliations":[{"raw_affiliation_string":"Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089622254","display_name":"Xuan Wang","orcid":"https://orcid.org/0000-0002-0830-0898"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuan Wang","raw_affiliation_strings":["Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)"],"affiliations":[{"raw_affiliation_string":"Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies,Harbin Institute of Technology (Shenzhen)","institution_ids":["https://openalex.org/I204983213"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5101188976"],"corresponding_institution_ids":["https://openalex.org/I204983213"],"apc_list":null,"apc_paid":null,"fwci":1.0848,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.81656446,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9689000248908997,"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"}},"topics":[{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9689000248908997,"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/T10551","display_name":"Scheduling and Optimization Algorithms","score":0.9595999717712402,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing 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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9502999782562256,"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":0.7995049953460693},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5890445113182068},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4543763995170593},{"id":"https://openalex.org/keywords/gpu-cluster","display_name":"GPU cluster","score":0.44210389256477356},{"id":"https://openalex.org/keywords/shared-resource","display_name":"Shared resource","score":0.43414872884750366},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3649117350578308},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.35893261432647705},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3577982485294342},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.22284382581710815},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.1860729455947876},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.07051724195480347}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7995049953460693},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5890445113182068},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4543763995170593},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.44210389256477356},{"id":"https://openalex.org/C51332947","wikidata":"https://www.wikidata.org/wiki/Q1172305","display_name":"Shared resource","level":2,"score":0.43414872884750366},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3649117350578308},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35893261432647705},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3577982485294342},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.22284382581710815},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.1860729455947876},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.07051724195480347},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/iwqos61813.2024.10682877","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iwqos61813.2024.10682877","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE/ACM 32nd International Symposium on Quality of Service (IWQoS)","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-156107","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-156107","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"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":"Conference paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W1512808804","https://openalex.org/W2034008996","https://openalex.org/W2074988668","https://openalex.org/W2335814492","https://openalex.org/W2760837370","https://openalex.org/W2798515322","https://openalex.org/W2918843973","https://openalex.org/W2919115771","https://openalex.org/W2919897868","https://openalex.org/W2920534947","https://openalex.org/W2963149401","https://openalex.org/W2982339048","https://openalex.org/W2982346115","https://openalex.org/W3012479151","https://openalex.org/W4221149636","https://openalex.org/W4297882618","https://openalex.org/W4318541676","https://openalex.org/W4383749535","https://openalex.org/W4394923413","https://openalex.org/W6684859321","https://openalex.org/W6756009870","https://openalex.org/W6758283263","https://openalex.org/W6758474236","https://openalex.org/W6759262305","https://openalex.org/W6759814162","https://openalex.org/W6760800931","https://openalex.org/W6769424276","https://openalex.org/W6774201734","https://openalex.org/W6781728138","https://openalex.org/W6782839094","https://openalex.org/W6799023841","https://openalex.org/W7067822191"],"related_works":["https://openalex.org/W2082176405","https://openalex.org/W2238531340","https://openalex.org/W2067902980","https://openalex.org/W2017587301","https://openalex.org/W1827514603","https://openalex.org/W2965037703","https://openalex.org/W2972376311","https://openalex.org/W1904802460","https://openalex.org/W3215286539","https://openalex.org/W2086392083"],"abstract_inverted_index":{"Deep":[0],"learning":[1],"(DL)":[2],"has":[3],"demonstrated":[4],"significant":[5],"success":[6],"across":[7],"diverse":[8],"fields,":[9],"leading":[10],"to":[11,33,89,110,158,195,234],"the":[12,83,112,187,196,206,213,224],"construction":[13],"of":[14,61,86,115],"dedicated":[15],"GPU":[16,19,62,87,146,226],"accelerators":[17],"within":[18],"clusters":[20,30],"for":[21,28,145,220],"high-quality":[22],"training":[23,120],"services.":[24],"Efficient":[25],"scheduler":[26],"designs":[27],"such":[29,68,211],"are":[31],"vital":[32],"reduce":[34,94],"operational":[35],"costs":[36],"and":[37,53,71,93,149,154,177,217],"enhance":[38],"resource":[39,91,147,208],"utilization.":[40],"While":[41],"recent":[42],"schedulers":[43],"have":[44],"shown":[45],"impressive":[46],"performance":[47,52,161],"in":[48,236],"optimizing":[49],"DL":[50,77,164,175,199],"job":[51,95,104,119,126,143,189],"cluster":[54],"utilization":[55,92],"through":[56,167],"periodic":[57],"reallocation":[58],"or":[59],"selection":[60],"resources,":[63],"they":[64],"also":[65],"encounter":[66],"challenges":[67],"as":[69,181,212],"preemption":[70],"migration":[72],"overhead,":[73],"along":[74],"with":[75,128,172],"potential":[76,84],"accuracy":[78,166],"degradation.":[79],"Nonetheless,":[80],"few":[81],"explore":[82],"benefits":[85],"sharing":[88,130,148,209,214,227],"improve":[90],"queuing":[96],"times.Motivated":[97],"by":[98,192,232],"these":[99],"insights,":[100],"we":[101],"present":[102],"a":[103,133,182],"scheduling":[105,138,155],"model":[106],"allowing":[107],"multiple":[108],"jobs":[109],"share":[111],"same":[113],"set":[114],"GPUs":[116],"without":[117],"altering":[118],"settings.":[121],"We":[122],"introduce":[123],"SJF-BSBF":[124,140,185,202],"(shortest":[125],"first":[127],"best":[129],"benefit":[131],"first),":[132],"straightforward":[134],"yet":[135],"effective":[136],"heuristic":[137],"algorithm.":[139],"intelligently":[141],"selects":[142],"pairs":[144],"runtime":[150],"settings":[151],"(sub-batch":[152],"size":[153,219],"time":[156,191,215],"point)":[157],"optimize":[159],"overall":[160],"while":[162],"ensuring":[163],"convergence":[165],"gradient":[168,221],"accumulation.":[169],"In":[170],"experiments":[171],"both":[173],"physical":[174],"workloads":[176],"trace-driven":[178],"simulations,":[179],"even":[180],"preemptionfree":[183],"policy,":[184],"reduces":[186],"average":[188],"completion":[190],"27-33%":[193],"relative":[194],"state-of-the-art":[197],"preemptive":[198],"schedulers.":[200],"Moreover,":[201],"can":[203],"wisely":[204],"determine":[205],"optimal":[207],"settings,":[210],"point":[216],"sub-batch":[218],"accumulation,":[222],"outperforming":[223],"aggressive":[225],"approach":[228],"(baseline":[229],"SJF-FFS":[230],"policy)":[231],"up":[233],"17%":[235],"large-scale":[237],"traces.":[238]},"counts_by_year":[{"year":2025,"cited_by_count":3}],"updated_date":"2026-03-12T08:34:05.389933","created_date":"2025-10-10T00:00:00"}