{"id":"https://openalex.org/W7225455368","doi":"https://doi.org/10.1922/comdem.v3i6s.18961","title":"Deep Reinforcement Learning for SLA-Aware GPU Scheduling Under Workload Uncertainty","display_name":"Deep Reinforcement Learning for SLA-Aware GPU Scheduling Under Workload Uncertainty","publication_year":2026,"publication_date":"2026-11-06","ids":{"openalex":"https://openalex.org/W7225455368","doi":"https://doi.org/10.1922/comdem.v3i6s.18961"},"language":"en","primary_location":{"id":"doi:10.1922/comdem.v3i6s.18961","doi":"https://doi.org/10.1922/comdem.v3i6s.18961","is_oa":true,"landing_page_url":"https://doi.org/10.1922/comdem.v3i6s.18961","pdf_url":"https://www.comdem.org/index.php/comdem/article/download/18961/12668","source":{"id":"https://openalex.org/S4404675604","display_name":"Computer and decision making.","issn_l":"3008-1416","issn":["3008-1416"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer and Decision Making: An International Journal","raw_type":"journal-article"},"volume":null,"type":"article","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://www.comdem.org/index.php/comdem/article/download/18961/12668","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5007333112","display_name":"Chetan N Madivalar","orcid":"https://orcid.org/0009-0000-1332-1899","observed_orcids":["https://orcid.org/0009-0000-1332-1899"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chetan N Madivalar","raw_affiliation_strings":["North American Lighting, Inc., Farmington Hills, Michigan, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"North American Lighting, Inc., Farmington Hills, Michigan, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089501861","display_name":"Yousef Alqasrawi","orcid":"https://orcid.org/0000-0001-9153-2085","observed_orcids":["https://orcid.org/0000-0001-9153-2085"]},"institutions":[{"id":"https://openalex.org/I146367977","display_name":"Applied Science Private University","ror":"https://ror.org/01ah6nb52","country_code":"JO","type":"education","lineage":["https://openalex.org/I146367977"]}],"countries":["JO"],"is_corresponding":false,"raw_author_name":"Yousef Alqasrawi","raw_affiliation_strings":["Computer Science, Faculty of Information Technology, Applied Science Private University, Amman, Jordan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Computer Science, Faculty of Information Technology, Applied Science Private University, Amman, Jordan","institution_ids":["https://openalex.org/I146367977"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"3","issue":"6S","first_page":"1990","last_page":"2010"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.5961999893188477,"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":0.5961999893188477,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.27469998598098755,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.06759999692440033,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/gpu-cluster-scheduling","display_name":"GPU cluster scheduling","score":0.9819999933242798},{"id":"https://openalex.org/keywords/service-level-agreements","display_name":"service level agreements","score":0.9160000085830688},{"id":"https://openalex.org/keywords/workload-uncertainty","display_name":"workload uncertainty","score":0.8799999952316284},{"id":"https://openalex.org/keywords/proximal-policy-optimization","display_name":"proximal policy optimization","score":0.8939999938011169},{"id":"https://openalex.org/keywords/deep-reinforcement-learning","display_name":"deep reinforcement learning","score":0.847000002861023},{"id":"https://openalex.org/keywords/queueing-delay","display_name":"queuing delay","score":0.9290000200271606},{"id":"https://openalex.org/keywords/resource-utilization","display_name":"resource utilization","score":0.9660000205039978},{"id":"https://openalex.org/keywords/action-masking","display_name":"action masking","score":0.878000020980835},{"id":"https://openalex.org/keywords/reward-function-design","display_name":"reward function design","score":0.640999972820282}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8090999722480774},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.7181000113487244},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6685000061988831},{"id":"https://openalex.org/C160403385","wikidata":"https://www.wikidata.org/wiki/Q220543","display_name":"Queue","level":2,"score":0.43369999527931213},{"id":"https://openalex.org/C107568181","wikidata":"https://www.wikidata.org/wiki/Q5319000","display_name":"Dynamic priority scheduling","level":3,"score":0.3874000012874603},{"id":"https://openalex.org/C32310161","wikidata":"https://www.wikidata.org/wiki/Q666365","display_name":"Earliest deadline first scheduling","level":5,"score":0.38690000772476196},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.38109999895095825},{"id":"https://openalex.org/C127456818","wikidata":"https://www.wikidata.org/wiki/Q238879","display_name":"Rate-monotonic scheduling","level":4,"score":0.36739999055862427},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.35830000042915344},{"id":"https://openalex.org/C171627638","wikidata":"https://www.wikidata.org/wiki/Q6206744","display_name":"Job queue","level":4,"score":0.3312999904155731},{"id":"https://openalex.org/C31689143","wikidata":"https://www.wikidata.org/wiki/Q733809","display_name":"Fair-share scheduling","level":3,"score":0.3278999924659729},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3160000145435333},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.31380000710487366},{"id":"https://openalex.org/C111873713","wikidata":"https://www.wikidata.org/wiki/Q1641413","display_name":"Job scheduler","level":3,"score":0.3125},{"id":"https://openalex.org/C119948110","wikidata":"https://www.wikidata.org/wiki/Q7858726","display_name":"Two-level scheduling","level":4,"score":0.30970001220703125},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.29980000853538513},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.28299999237060547},{"id":"https://openalex.org/C122141398","wikidata":"https://www.wikidata.org/wiki/Q5456330","display_name":"Fixed-priority pre-emptive scheduling","level":5,"score":0.2768000066280365},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2728999853134155},{"id":"https://openalex.org/C202532154","wikidata":"https://www.wikidata.org/wiki/Q4374193","display_name":"Problem statement","level":2,"score":0.25600001215934753}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1922/comdem.v3i6s.18961","doi":"https://doi.org/10.1922/comdem.v3i6s.18961","is_oa":true,"landing_page_url":"https://doi.org/10.1922/comdem.v3i6s.18961","pdf_url":"https://www.comdem.org/index.php/comdem/article/download/18961/12668","source":{"id":"https://openalex.org/S4404675604","display_name":"Computer and decision making.","issn_l":"3008-1416","issn":["3008-1416"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer and Decision Making: An International Journal","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1922/comdem.v3i6s.18961","doi":"https://doi.org/10.1922/comdem.v3i6s.18961","is_oa":true,"landing_page_url":"https://doi.org/10.1922/comdem.v3i6s.18961","pdf_url":"https://www.comdem.org/index.php/comdem/article/download/18961/12668","source":{"id":"https://openalex.org/S4404675604","display_name":"Computer and decision making.","issn_l":"3008-1416","issn":["3008-1416"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer and Decision Making: An International Journal","raw_type":"journal-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7225455368.pdf","grobid_xml":"https://content.openalex.org/works/W7225455368.grobid-xml"},"referenced_works_count":22,"referenced_works":["https://openalex.org/W3038822267","https://openalex.org/W3139398491","https://openalex.org/W4318541676","https://openalex.org/W4367311517","https://openalex.org/W4372262787","https://openalex.org/W4380325188","https://openalex.org/W4382196419","https://openalex.org/W4382246105","https://openalex.org/W4390263770","https://openalex.org/W4390939129","https://openalex.org/W4392196196","https://openalex.org/W4400905293","https://openalex.org/W4401954787","https://openalex.org/W4402570177","https://openalex.org/W4403311596","https://openalex.org/W4404385885","https://openalex.org/W4410197948","https://openalex.org/W4411403346","https://openalex.org/W4413395561","https://openalex.org/W7125601066","https://openalex.org/W7169808147","https://openalex.org/W7203756369"],"related_works":["https://openalex.org/W7169808147","https://openalex.org/W7160170996","https://openalex.org/W4392309097","https://openalex.org/W2975995881","https://openalex.org/W4392699261","https://openalex.org/W7125601066","https://openalex.org/W4393186158","https://openalex.org/W7171274090","https://openalex.org/W3010528756","https://openalex.org/W7037271142"],"abstract_inverted_index":{"GPU":[0],"cluster":[1],"scheduling":[2,32,144,167],"involves":[3],"making":[4,207],"choices":[5],"to":[6,72,81],"meet":[7],"service-level":[8],"agreement":[9],"(SLA)":[10],"goals":[11],"while":[12],"minimizing":[13],"queue":[14],"delay":[15],"and":[16,51,76,101,104,118,126,137,209,223],"maximizing":[17],"resource":[18],"utilization":[19,106],"when":[20],"jobs":[21],"have":[22],"variable":[23],"runtimes.":[24],"In":[25],"this":[26],"paper,":[27],"we":[28],"introduce":[29],"an":[30,120,141,210],"SLA-aware":[31],"framework,":[33],"using":[34],"proximal":[35],"policy":[36],"optimization":[37],"(PPO).":[38],"The":[39,108,146],"proposed":[40],"observation":[41],"is":[42,79,93,163,201],"a":[43,52,94,105,123,133,151,187,202,205],"nine":[44],"dimensional":[45],"vector":[46],"in":[47,176],"the":[48,115,157,177,181,214],"global":[49],"space":[50],"tensor":[53],"of":[54,63,84,96,122,135,195,213],"candidate":[55,65],"job":[56],"features,":[57],"with":[58],"thirteen":[59],"features":[60],"for":[61,98,114,189,204,217],"each":[62],"ten":[64],"jobs.":[66],"Historical":[67],"runtime":[68,124],"summaries/histograms":[69],"are":[70,87,227],"provided":[71,80,175],"indicate":[73],"duration":[74],"variability,":[75],"action":[77],"masking":[78],"exclude":[82],"selections":[83],"resources":[85],"that":[86],"not":[88,140,185],"feasible.":[89],"A":[90],"weighted":[91],"reward":[92],"combination":[95],"penalties":[97],"queue-threshold":[99],"violations":[100],"waiting":[102],"time,":[103],"incentive.":[107],"following":[109],"notebook":[110],"contains":[111],"pre-defined":[112],"values":[113],"scheduler":[116],"comparisons":[117,183],"ablation,":[119],"example":[121],"distribution":[125],"scripted":[127],"episode":[128],"sequences.":[129],"These":[130],"materials":[131],"show":[132],"way":[134],"presenting":[136],"visualizing":[138],"numbers,":[139],"actually":[142],"conducted":[143],"experiment.":[145],"predefined":[147],"PPO":[148,169],"entry":[149],"has":[150],"lower":[152],"SLA":[153],"violation":[154],"value":[155],"than":[156],"listed":[158],"comparison":[159],"entries,":[160],"but":[161],"there":[162],"no":[164],"trace-loading":[165],"pipeline,":[166],"simulator,":[168],"implementation":[170],"or":[171,192,197],"independent":[172],"evaluation":[173],"logs":[174],"supplied":[178],"notebook.":[179],"Therefore,":[180],"numerical":[182],"do":[184],"provide":[186],"basis":[188],"learned-policy":[190],"superiority":[191],"convergence,":[193],"effectiveness":[194],"uncertainty-models,":[196],"production-trace":[198],"performance.":[199],"This":[200],"proposal":[203],"decision":[206],"framework":[208],"explicit":[211],"statement":[212],"evidence":[215],"needed":[216],"evaluation.":[218],"Empirical":[219],"validation,":[220],"sensitivity":[221],"testing":[222],"repeated":[224],"run":[225],"assessment":[226],"future":[228],"works.":[229]},"counts_by_year":[],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2026-11-08T00:00:00"}