{"id":"https://openalex.org/W4394896940","doi":"https://doi.org/10.1109/tpds.2024.3390109","title":"Sampling-Based Multi-Job Placement for Heterogeneous Deep Learning Clusters","display_name":"Sampling-Based Multi-Job Placement for Heterogeneous Deep Learning Clusters","publication_year":2024,"publication_date":"2024-04-17","ids":{"openalex":"https://openalex.org/W4394896940","doi":"https://doi.org/10.1109/tpds.2024.3390109"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2024.3390109","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2024.3390109","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/A5024284822","display_name":"Kaiyang Liu","orcid":"https://orcid.org/0000-0002-1114-8030"},"institutions":[{"id":"https://openalex.org/I130438778","display_name":"Memorial University of Newfoundland","ror":"https://ror.org/04haebc03","country_code":"CA","type":"education","lineage":["https://openalex.org/I130438778"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Kaiyang Liu","raw_affiliation_strings":["Department of Computer Science, Memorial University of Newfoundland, St. John’s, NL, Canada"],"raw_orcid":"https://orcid.org/0000-0002-1114-8030","affiliations":[{"raw_affiliation_string":"Department of Computer Science, Memorial University of Newfoundland, St. John’s, NL, Canada","institution_ids":["https://openalex.org/I130438778"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101992293","display_name":"Jingrong Wang","orcid":"https://orcid.org/0000-0003-3392-7016"},"institutions":[{"id":"https://openalex.org/I185261750","display_name":"University of Toronto","ror":"https://ror.org/03dbr7087","country_code":"CA","type":"education","lineage":["https://openalex.org/I185261750"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Jingrong Wang","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON, Canada"],"raw_orcid":"https://orcid.org/0000-0003-3392-7016","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON, Canada","institution_ids":["https://openalex.org/I185261750"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022175140","display_name":"Zhiming Huang","orcid":"https://orcid.org/0000-0001-6465-591X"},"institutions":[{"id":"https://openalex.org/I212119943","display_name":"University of Victoria","ror":"https://ror.org/04s5mat29","country_code":"CA","type":"education","lineage":["https://openalex.org/I212119943"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Zhiming Huang","raw_affiliation_strings":["Department of Computer Science, University of Victoria, Victoria, BC, Canada"],"raw_orcid":"https://orcid.org/0000-0001-6465-591X","affiliations":[{"raw_affiliation_string":"Department of Computer Science, University of Victoria, Victoria, BC, Canada","institution_ids":["https://openalex.org/I212119943"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5043097022","display_name":"Jianping Pan","orcid":"https://orcid.org/0000-0003-4893-6847"},"institutions":[{"id":"https://openalex.org/I212119943","display_name":"University of Victoria","ror":"https://ror.org/04s5mat29","country_code":"CA","type":"education","lineage":["https://openalex.org/I212119943"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Jianping Pan","raw_affiliation_strings":["Department of Computer Science, University of Victoria, Victoria, BC, Canada"],"raw_orcid":"https://orcid.org/0000-0003-4893-6847","affiliations":[{"raw_affiliation_string":"Department of Computer Science, University of Victoria, Victoria, BC, Canada","institution_ids":["https://openalex.org/I212119943"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6808,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.85364305,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":"35","issue":"6","first_page":"1029","last_page":"1043"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9993000030517578,"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":0.9993000030517578,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.9976999759674072,"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":0.9976000189781189,"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":0.8373482823371887},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.6282826662063599},{"id":"https://openalex.org/keywords/job-scheduler","display_name":"Job scheduler","score":0.562214732170105},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5411758422851562},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.5259174108505249},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4594295024871826},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4551837146282196},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.41903120279312134},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.2726154923439026},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.17018145322799683}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8373482823371887},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.6282826662063599},{"id":"https://openalex.org/C111873713","wikidata":"https://www.wikidata.org/wiki/Q1641413","display_name":"Job scheduler","level":3,"score":0.562214732170105},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5411758422851562},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5259174108505249},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4594295024871826},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4551837146282196},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.41903120279312134},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2726154923439026},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.17018145322799683},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C160403385","wikidata":"https://www.wikidata.org/wiki/Q220543","display_name":"Queue","level":2,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2024.3390109","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2024.3390109","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":"Decent work and economic growth","score":0.6899999976158142,"id":"https://metadata.un.org/sdg/8"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320319952","display_name":"Canada Foundation for Innovation","ror":"https://ror.org/000az4664"},{"id":"https://openalex.org/F4320334593","display_name":"Natural Sciences and Engineering Research Council of Canada","ror":"https://ror.org/01h531d29"},{"id":"https://openalex.org/F4320335271","display_name":"British Columbia Knowledge Development Fund","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W186403072","https://openalex.org/W1630733292","https://openalex.org/W2013683781","https://openalex.org/W2105947650","https://openalex.org/W2155893237","https://openalex.org/W2169384417","https://openalex.org/W2186615578","https://openalex.org/W2194775991","https://openalex.org/W2523435939","https://openalex.org/W2798515322","https://openalex.org/W2896457183","https://openalex.org/W2919897868","https://openalex.org/W2946058832","https://openalex.org/W3012220622","https://openalex.org/W3047783725","https://openalex.org/W3142588439","https://openalex.org/W3189640247","https://openalex.org/W3203600045","https://openalex.org/W4211136162","https://openalex.org/W4226160060","https://openalex.org/W4236491543","https://openalex.org/W4238465620","https://openalex.org/W4281805071","https://openalex.org/W4392454234","https://openalex.org/W6636933019","https://openalex.org/W6679393576","https://openalex.org/W6680402377","https://openalex.org/W6686509673","https://openalex.org/W6713134421","https://openalex.org/W6736413256","https://openalex.org/W6753584990","https://openalex.org/W6755207826","https://openalex.org/W6759814162","https://openalex.org/W6769424276","https://openalex.org/W6775649616","https://openalex.org/W6782839094","https://openalex.org/W6784425352","https://openalex.org/W6784981981","https://openalex.org/W6839007396","https://openalex.org/W6845510308"],"related_works":["https://openalex.org/W1582214068","https://openalex.org/W44553394","https://openalex.org/W2172100609","https://openalex.org/W3174084021","https://openalex.org/W1970208768","https://openalex.org/W1559381305","https://openalex.org/W3047653192","https://openalex.org/W4283696703","https://openalex.org/W381368390","https://openalex.org/W1984090128"],"abstract_inverted_index":{"Heterogeneous":[0],"deep":[1,80,145],"learning":[2,10,19,34,81,188],"clusters":[3],"commonly":[4],"host":[5],"a":[6,54],"variety":[7],"of":[8,18,79,123,140,171],"distributed":[9],"jobs.":[11,82,189],"In":[12],"such":[13],"scenarios,":[14],"the":[15,25,30,59,73,106,121,138,161,172],"training":[16,31,89],"efficiency":[17],"models":[20],"is":[21,98],"negatively":[22],"affected":[23],"by":[24],"slowest":[26],"worker.":[27],"To":[28,103],"accelerate":[29],"process,":[32],"multiple":[33],"jobs":[35],"may":[36],"compete":[37],"for":[38,109],"limited":[39],"computational":[40,107],"resources,":[41],"posing":[42],"significant":[43],"challenges":[44],"to":[45,57,119,158],"multi-job":[46,60],"placement":[47,61],"among":[48,187],"heterogeneous":[49,150],"workers.":[50],"This":[51],"paper":[52],"presents":[53],"heterogeneity-aware":[55],"scheduler":[56],"solve":[58],"problem":[62],"while":[63],"taking":[64],"into":[65],"account":[66],"job":[67],"sizing":[68],"and":[69,95,113,182],"load":[70],"balancing,":[71],"minimizing":[72],"average":[74],"Job":[75],"Completion":[76],"Time":[77],"(JCT)":[78],"A":[83],"novel":[84],"scheme":[85,164],"based":[86],"on":[87,149],"proportional":[88],"workload":[90],"assignment,":[91],"feasible":[92,124],"solution":[93,125],"categorization,":[94],"matching":[96],"markets":[97],"proposed":[99,162],"with":[100,143,175],"theoretical":[101],"guarantees.":[102],"further":[104],"reduce":[105],"complexity":[108],"low":[110],"latency":[111],"decision-making":[112],"improve":[114],"scheduling":[115,185],"fairness,":[116],"we":[117],"propose":[118],"construct":[120],"sparsification":[122],"categories":[126],"through":[127],"sampling,":[128],"which":[129],"has":[130],"negligible":[131],"performance":[132,139],"loss":[133],"in":[134,178],"JCT.":[135],"We":[136],"evaluate":[137],"our":[141],"design":[142],"real-world":[144],"neural":[146],"network":[147],"benchmarks":[148],"computing":[151],"clusters.":[152],"Experimental":[153],"results":[154,168],"show":[155],"that,":[156],"compared":[157],"existing":[159],"solutions,":[160],"sampling-based":[163],"can":[165],"achieve":[166],"1)":[167],"within":[169],"2.04%":[170],"optimal":[173],"JCT":[174],"orders-of-magnitude":[176],"improvements":[177],"algorithm":[179],"running":[180],"time,":[181],"2)":[183],"high":[184],"fairness":[186]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":2}],"updated_date":"2026-07-29T09:40:50.615796","created_date":"2025-10-10T00:00:00"}