{"id":"https://openalex.org/W4416933662","doi":"https://doi.org/10.1002/cpe.70456","title":"Dynamic Load Balancing for Distributed Large Model Training: A Hybrid Framework of Gray Markov Chain and <scp>MDP</scp>","display_name":"Dynamic Load Balancing for Distributed Large Model Training: A Hybrid Framework of Gray Markov Chain and <scp>MDP</scp>","publication_year":2025,"publication_date":"2025-12-02","ids":{"openalex":"https://openalex.org/W4416933662","doi":"https://doi.org/10.1002/cpe.70456"},"language":"en","primary_location":{"id":"doi:10.1002/cpe.70456","is_oa":false,"landing_page_url":"https://doi.org/10.1002/cpe.70456","pdf_url":null,"source":{"id":"https://openalex.org/S11065456","display_name":"Concurrency and Computation Practice and Experience","issn_l":"1532-0626","issn":["1532-0626","1532-0634"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320595","host_organization_name":"Wiley","host_organization_lineage":["https://openalex.org/P4310320595"],"host_organization_lineage_names":["Wiley"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Concurrency and Computation: Practice and Experience","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/A5100364659","display_name":"Yonggang Li","orcid":"https://orcid.org/0009-0005-1836-2450"},"institutions":[{"id":"https://openalex.org/I10535382","display_name":"Chongqing University of Posts and Telecommunications","ror":"https://ror.org/03dgaqz26","country_code":"CN","type":"education","lineage":["https://openalex.org/I10535382"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yonggang Li","raw_affiliation_strings":["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"],"raw_orcid":"https://orcid.org/0009-0005-1836-2450","affiliations":[{"raw_affiliation_string":"School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China","institution_ids":["https://openalex.org/I10535382"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082622488","display_name":"Rui Ji","orcid":"https://orcid.org/0000-0003-2920-5766"},"institutions":[{"id":"https://openalex.org/I10535382","display_name":"Chongqing University of Posts and Telecommunications","ror":"https://ror.org/03dgaqz26","country_code":"CN","type":"education","lineage":["https://openalex.org/I10535382"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rui Ji","raw_affiliation_strings":["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China","institution_ids":["https://openalex.org/I10535382"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101268951","display_name":"Yaotong Su","orcid":null},"institutions":[{"id":"https://openalex.org/I10535382","display_name":"Chongqing University of Posts and Telecommunications","ror":"https://ror.org/03dgaqz26","country_code":"CN","type":"education","lineage":["https://openalex.org/I10535382"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yaotong Su","raw_affiliation_strings":["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China","institution_ids":["https://openalex.org/I10535382"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115593972","display_name":"Yuanjin Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I10535382","display_name":"Chongqing University of Posts and Telecommunications","ror":"https://ror.org/03dgaqz26","country_code":"CN","type":"education","lineage":["https://openalex.org/I10535382"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuanjin Zhang","raw_affiliation_strings":["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China","institution_ids":["https://openalex.org/I10535382"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061523006","display_name":"Andong Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I10535382","display_name":"Chongqing University of Posts and Telecommunications","ror":"https://ror.org/03dgaqz26","country_code":"CN","type":"education","lineage":["https://openalex.org/I10535382"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Andong Zhang","raw_affiliation_strings":["School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Communications and Information Engineering Chongqing University of Posts and Telecommunications  Chongqing China","institution_ids":["https://openalex.org/I10535382"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5103281102","display_name":"Longjiang Li","orcid":"https://orcid.org/0000-0003-1388-3300"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]},{"id":"https://openalex.org/I24201400","display_name":"Chengdu University of Information Technology","ror":"https://ror.org/01yxwrh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I24201400"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Longjiang Li","raw_affiliation_strings":["School of Information and Communication Engineering University of Electronic Science and Technology  Chengdu China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information and Communication Engineering University of Electronic Science and Technology  Chengdu China","institution_ids":["https://openalex.org/I150229711","https://openalex.org/I24201400"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5100364659"],"corresponding_institution_ids":["https://openalex.org/I10535382"],"apc_list":{"value":4740,"currency":"USD","value_usd":4740},"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.38455974,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"38","issue":"1","first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10714","display_name":"Software-Defined Networks and 5G","score":0.35109999775886536,"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/T10714","display_name":"Software-Defined Networks and 5G","score":0.35109999775886536,"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":0.25450000166893005,"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/T12127","display_name":"Software System Performance and Reliability","score":0.07339999824762344,"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/scalability","display_name":"Scalability","score":0.7019000053405762},{"id":"https://openalex.org/keywords/load-balancing","display_name":"Load balancing (electrical power)","score":0.5253000259399414},{"id":"https://openalex.org/keywords/markov-chain","display_name":"Markov chain","score":0.5156999826431274},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.450300008058548},{"id":"https://openalex.org/keywords/data-transmission","display_name":"Data transmission","score":0.3993000090122223},{"id":"https://openalex.org/keywords/dynamic-network-analysis","display_name":"Dynamic network analysis","score":0.3930000066757202},{"id":"https://openalex.org/keywords/resource-allocation","display_name":"Resource allocation","score":0.3601999878883362},{"id":"https://openalex.org/keywords/markov-process","display_name":"Markov process","score":0.35839998722076416},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.3555999994277954}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.90829998254776},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7019000053405762},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5928000211715698},{"id":"https://openalex.org/C138959212","wikidata":"https://www.wikidata.org/wiki/Q1806783","display_name":"Load balancing (electrical power)","level":3,"score":0.5253000259399414},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.5156999826431274},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.450300008058548},{"id":"https://openalex.org/C557945733","wikidata":"https://www.wikidata.org/wiki/Q389772","display_name":"Data transmission","level":2,"score":0.3993000090122223},{"id":"https://openalex.org/C13540734","wikidata":"https://www.wikidata.org/wiki/Q5318996","display_name":"Dynamic network analysis","level":2,"score":0.3930000066757202},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.3601999878883362},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.35839998722076416},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.3555999994277954},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.31690001487731934},{"id":"https://openalex.org/C203274722","wikidata":"https://www.wikidata.org/wiki/Q7001161","display_name":"Network performance","level":2,"score":0.31630000472068787},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.31290000677108765},{"id":"https://openalex.org/C2779370713","wikidata":"https://www.wikidata.org/wiki/Q357554","display_name":"Load management","level":2,"score":0.30820000171661377},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.29910001158714294},{"id":"https://openalex.org/C163836022","wikidata":"https://www.wikidata.org/wiki/Q6771326","display_name":"Markov model","level":3,"score":0.2842000126838684},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.28299999237060547},{"id":"https://openalex.org/C107568181","wikidata":"https://www.wikidata.org/wiki/Q5319000","display_name":"Dynamic priority scheduling","level":3,"score":0.28029999136924744},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2770000100135803},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.2703999876976013},{"id":"https://openalex.org/C19012869","wikidata":"https://www.wikidata.org/wiki/Q578372","display_name":"Response time","level":2,"score":0.2671999931335449},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.26100000739097595},{"id":"https://openalex.org/C127964446","wikidata":"https://www.wikidata.org/wiki/Q1092142","display_name":"Computational resource","level":3,"score":0.25600001215934753}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1002/cpe.70456","is_oa":false,"landing_page_url":"https://doi.org/10.1002/cpe.70456","pdf_url":null,"source":{"id":"https://openalex.org/S11065456","display_name":"Concurrency and Computation Practice and Experience","issn_l":"1532-0626","issn":["1532-0626","1532-0634"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320595","host_organization_name":"Wiley","host_organization_lineage":["https://openalex.org/P4310320595"],"host_organization_lineage_names":["Wiley"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Concurrency and Computation: Practice and Experience","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G7792699363","display_name":null,"funder_award_id":"2022YFC3005702","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W2060204507","https://openalex.org/W2963341956","https://openalex.org/W2997200074","https://openalex.org/W3034674511","https://openalex.org/W3086105743","https://openalex.org/W3107548300","https://openalex.org/W3204998121","https://openalex.org/W4205946471","https://openalex.org/W4302018437","https://openalex.org/W4309696447","https://openalex.org/W4312981186","https://openalex.org/W4313555487","https://openalex.org/W4388236136","https://openalex.org/W4388819905","https://openalex.org/W4388856373","https://openalex.org/W4390970405","https://openalex.org/W4392941142","https://openalex.org/W4396784431","https://openalex.org/W4401508395","https://openalex.org/W4401834466","https://openalex.org/W4401929513","https://openalex.org/W4402670757","https://openalex.org/W4404628621","https://openalex.org/W4405270867"],"related_works":[],"abstract_inverted_index":{"ABSTRACT":[0],"Large\u2010scale":[1],"model":[2,67,97,113],"training":[3,170,198],"in":[4,12,166,172,200],"distributed":[5,169,197],"data":[6,65,115,134],"centers":[7],"plays":[8],"a":[9,75],"crucial":[10],"role":[11],"deep":[13,182],"learning.":[14],"Still,":[15],"it":[16],"faces":[17],"significant":[18],"challenges,":[19,56],"including":[20],"resource":[21,131,147],"fragmentation,":[22],"low":[23],"bandwidth":[24],"utilization,":[25,148],"and":[26,50,69,80,91,133,149,179,194],"complex":[27],"task":[28],"flow":[29,135],"management.":[30,122],"The":[31,95,123,156],"problem":[32],"is":[33,98],"exacerbated":[34],"by":[35],"high\u2010speed,":[36],"high\u2010capacity":[37],"parameter":[38],"synchronization,":[39],"often":[40],"exceeding":[41],"several":[42],"hundred":[43],"Gbps,":[44],"which":[45],"leads":[46],"to":[47,100,118,153],"reduced":[48],"throughput":[49],"computational":[51,93,150],"inefficiencies.":[52],"To":[53],"address":[54],"these":[55,126],"this":[57,141,160],"paper":[58],"proposes":[59],"an":[60],"innovative":[61],"approach":[62,87,143,162],"that":[63,140,159],"combines":[64],"parallelism,":[66,68],"dynamic":[70],"load":[71],"balancing.":[72],"By":[73],"integrating":[74],"Gray":[76],"Markov":[77,81],"Chain":[78],"(GMC)":[79],"Decision":[82],"Process":[83],"(MDP)":[84],"model,":[85],"the":[86,111,177,192],"dynamically":[88],"schedules":[89],"resources":[90],"balances":[92],"loads.":[94],"GMC":[96],"used":[99],"predict":[101],"future":[102],"node":[103],"loads,":[104],"facilitating":[105],"optimal":[106],"weight":[107],"matrix":[108],"decomposition,":[109],"while":[110],"MDP":[112],"adjusts":[114],"transmission":[116],"paths":[117],"optimize":[119],"network":[120],"traffic":[121],"combination":[124],"of":[125,181,196],"two":[127],"models":[128],"enhances":[129],"both":[130],"allocation":[132],"optimization.":[136],"Experimental":[137],"results":[138],"demonstrate":[139],"integrated":[142],"significantly":[144,175],"improves":[145],"throughput,":[146],"efficiency":[151,193],"compared":[152],"traditional":[154],"methods.":[155],"findings":[157],"suggest":[158],"hybrid":[161],"performs":[163],"exceptionally":[164],"well":[165],"optimizing":[167],"large\u2010scale":[168],"tasks":[171],"multidata\u2010center":[173],"environments,":[174],"improving":[176],"scalability":[178],"performance":[180],"learning":[183],"workloads.":[184],"This":[185],"research":[186],"shows":[187],"promising":[188],"implications":[189],"for":[190],"enhancing":[191],"effectiveness":[195],"systems":[199],"high\u2010demand":[201],"applications.":[202]},"counts_by_year":[],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-12-03T00:00:00"}
