{"id":"https://openalex.org/W4385566059","doi":"https://doi.org/10.1145/3580305.3599241","title":"A Dual-Agent Scheduler for Distributed Deep Learning Jobs on Public Cloud via Reinforcement Learning","display_name":"A Dual-Agent Scheduler for Distributed Deep Learning Jobs on Public Cloud via Reinforcement Learning","publication_year":2023,"publication_date":"2023-08-04","ids":{"openalex":"https://openalex.org/W4385566059","doi":"https://doi.org/10.1145/3580305.3599241"},"language":"en","primary_location":{"id":"doi:10.1145/3580305.3599241","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3580305.3599241","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","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/A5060626806","display_name":"Mingzhe Xing","orcid":"https://orcid.org/0000-0002-2065-9852"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Mingzhe Xing","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076505011","display_name":"Hangyu Mao","orcid":"https://orcid.org/0000-0002-4499-7581"},"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":"Hangyu Mao","raw_affiliation_strings":["Sensetime Research, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Sensetime Research, Beijing, China","institution_ids":["https://openalex.org/I4210128910"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001804019","display_name":"Shenglin Yin","orcid":"https://orcid.org/0000-0002-5216-9946"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shenglin Yin","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091272448","display_name":"Lichen Pan","orcid":"https://orcid.org/0000-0001-7451-0140"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lichen Pan","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101471395","display_name":"Zhengchao Zhang","orcid":"https://orcid.org/0009-0002-4782-9046"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhengchao Zhang","raw_affiliation_strings":["ByteDance, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"ByteDance, Shanghai, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102979232","display_name":"Zhen Xiao","orcid":"https://orcid.org/0000-0002-6784-9709"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhen Xiao","raw_affiliation_strings":["Peking University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5017295101","display_name":"Jieyi Long","orcid":"https://orcid.org/0009-0007-4646-7131"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jieyi Long","raw_affiliation_strings":["Theta Labs, Inc., San Jose, USA"],"affiliations":[{"raw_affiliation_string":"Theta Labs, Inc., San Jose, USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5060626806"],"corresponding_institution_ids":["https://openalex.org/I20231570"],"apc_list":null,"apc_paid":null,"fwci":1.4064,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.82320154,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"2776","last_page":"2788"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9987000226974487,"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"}},"topics":[{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9987000226974487,"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.9980999827384949,"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":0.9937999844551086,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8590068817138672},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.8369633555412292},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.6771417856216431},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.6479426622390747},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.6046604514122009},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5668460130691528},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4661855399608612},{"id":"https://openalex.org/keywords/job-scheduler","display_name":"Job scheduler","score":0.46615397930145264},{"id":"https://openalex.org/keywords/automated-planning-and-scheduling","display_name":"Automated planning and scheduling","score":0.418326735496521},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3613660931587219},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.1346360743045807},{"id":"https://openalex.org/keywords/database","display_name":"Database","score":0.12464874982833862}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8590068817138672},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8369633555412292},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.6771417856216431},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.6479426622390747},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.6046604514122009},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5668460130691528},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4661855399608612},{"id":"https://openalex.org/C111873713","wikidata":"https://www.wikidata.org/wiki/Q1641413","display_name":"Job scheduler","level":3,"score":0.46615397930145264},{"id":"https://openalex.org/C114073186","wikidata":"https://www.wikidata.org/wiki/Q2631895","display_name":"Automated planning and scheduling","level":2,"score":0.418326735496521},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3613660931587219},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.1346360743045807},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.12464874982833862},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3580305.3599241","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3580305.3599241","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/8","score":0.4399999976158142,"display_name":"Decent work and economic growth"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W95608104","https://openalex.org/W1502922572","https://openalex.org/W1567012231","https://openalex.org/W2037801351","https://openalex.org/W2057913648","https://openalex.org/W2088805718","https://openalex.org/W2098126897","https://openalex.org/W2100830825","https://openalex.org/W2102959510","https://openalex.org/W2105947650","https://openalex.org/W2111038936","https://openalex.org/W2117231097","https://openalex.org/W2129375853","https://openalex.org/W2153070348","https://openalex.org/W2194775991","https://openalex.org/W2407879741","https://openalex.org/W2488653541","https://openalex.org/W2492608700","https://openalex.org/W2546571074","https://openalex.org/W2604272474","https://openalex.org/W2617547828","https://openalex.org/W2884700152","https://openalex.org/W2996834744","https://openalex.org/W2998489261","https://openalex.org/W3011330593","https://openalex.org/W3013664888","https://openalex.org/W3036265301","https://openalex.org/W3037847693","https://openalex.org/W3088310808","https://openalex.org/W3120778962","https://openalex.org/W3121689374","https://openalex.org/W3153881663","https://openalex.org/W3162118826","https://openalex.org/W3167705651","https://openalex.org/W3172874292","https://openalex.org/W3197816522","https://openalex.org/W3206254925","https://openalex.org/W3209503812","https://openalex.org/W4205916020","https://openalex.org/W4206174365","https://openalex.org/W4214717370","https://openalex.org/W4232967792","https://openalex.org/W4285604331","https://openalex.org/W4294982491","https://openalex.org/W4306679387","https://openalex.org/W6758383543"],"related_works":["https://openalex.org/W3033920833","https://openalex.org/W3038948847","https://openalex.org/W44553394","https://openalex.org/W2121106196","https://openalex.org/W2189586021","https://openalex.org/W4382866894","https://openalex.org/W4234499367","https://openalex.org/W2755738839","https://openalex.org/W2116158592","https://openalex.org/W1972259977"],"abstract_inverted_index":{"Public":[0],"cloud":[1],"GPU":[2,84,215],"clusters":[3],"are":[4,71,105],"becoming":[5],"emerging":[6],"platforms":[7],"for":[8,43,73,194,197],"training":[9,16,31],"distributed":[10],"deep":[11],"learning":[12],"jobs.":[13],"Under":[14],"this":[15,158],"paradigm,":[17],"the":[18,47,75,89,132,167,173,198,211,221,226,239,245],"job":[19,34,145],"scheduler":[20,163],"is":[21,50,228],"a":[22,161,190,203],"crucial":[23],"component":[24],"to":[25,52,88,170,209],"improve":[26],"user":[27],"experiences,":[28],"i.e.,":[29,64],"reducing":[30],"fees":[32],"and":[33,67,80,136,142,175,178],"completion":[35],"time,":[36],"which":[37,70],"can":[38,95],"also":[39],"save":[40],"power":[41],"costs":[42],"service":[44],"providers.":[45],"However,":[46],"scheduling":[48,76,181],"problem":[49],"known":[51],"be":[53,149],"NP-hard.":[54],"Most":[55],"existing":[56,152],"work":[57],"divides":[58],"it":[59],"into":[60],"two":[61,107,168],"easier":[62],"sub-tasks,":[63],"ordering":[65,122,174,187],"task":[66],"placement":[68,81,124,176,199],"task,":[69],"responsible":[72],"deciding":[74],"orders":[77,82],"of":[78,83,214,220,247],"jobs":[79],"machines,":[85],"respectively.":[86],"Due":[87],"superior":[90],"adaptation":[91],"ability,":[92],"learning-based":[93,117],"policies":[94,177],"generally":[96],"perform":[97],"better":[98,195],"than":[99],"traditional":[100],"heuristic-based":[101],"methods.":[102],"Nevertheless,":[103],"there":[104],"still":[106],"main":[108],"challenges":[109],"that":[110],"have":[111],"not":[112],"been":[113],"well-solved.":[114],"First,":[115],"most":[116],"methods":[118],"only":[119],"focus":[120],"on":[121,144,238],"or":[123],"policy":[125],"independently,":[126],"while":[127,217],"ignoring":[128],"their":[129],"cooperation.":[130],"Second,":[131],"unbalanced":[133],"machine":[134],"performances":[135],"resource":[137],"contention":[138],"impose":[139],"huge":[140],"overhead":[141],"uncertainty":[143],"duration,":[146],"but":[147],"rarely":[148],"considered":[150],"in":[151],"work.":[153],"To":[154],"tackle":[155],"these":[156],"issues,":[157],"paper":[159],"presents":[160],"dual-agent":[162,227],"framework":[164],"abstracted":[165],"from":[166],"sub-tasks":[169],"jointly":[171,229],"learn":[172,210],"make":[179],"better-informed":[180],"decisions.":[182],"Specifically,":[183],"we":[184,201],"design":[185],"an":[186],"agent":[188],"with":[189,231],"scalable":[191],"squeeze-and-communicate":[192],"strategy":[193],"cooperation;":[196],"agent,":[200],"propose":[202],"novel":[204],"Random":[205],"Walk":[206],"Gaussian":[207],"Process":[208],"performance":[212,223],"similarities":[213],"machines":[216],"being":[218],"aware":[219],"uncertain":[222],"fluctuation.":[224],"Finally,":[225],"optimized":[230],"multi-agent":[232],"reinforcement":[233],"learning.":[234],"Extensive":[235],"experiments":[236],"conducted":[237],"real-world":[240],"production":[241],"cluster":[242],"trace":[243],"demonstrate":[244],"superiority":[246],"our":[248],"model.":[249]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":4}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}