{"id":"https://openalex.org/W7155508889","doi":"https://doi.org/10.1145/3767295.3803623","title":"Suika: Efficient and High-quality Re-scheduling of 3D-parallelized LLM Training Jobs in Shared Clusters","display_name":"Suika: Efficient and High-quality Re-scheduling of 3D-parallelized LLM Training Jobs in Shared Clusters","publication_year":2026,"publication_date":"2026-04-24","ids":{"openalex":"https://openalex.org/W7155508889","doi":"https://doi.org/10.1145/3767295.3803623"},"language":null,"primary_location":{"id":"doi:10.1145/3767295.3803623","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767295.3803623","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st European Conference on Computer Systems","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3767295.3803623","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134544287","display_name":"Yuxuan Wang","orcid":"https://orcid.org/0009-0001-1595-411X"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I4210136246","display_name":"China Telecom (China)","ror":"https://ror.org/03jgnzt20","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210136246"]},{"id":"https://openalex.org/I4387153335","display_name":"China Telecom","ror":"https://ror.org/05p67dv18","country_code":null,"type":"company","lineage":["https://openalex.org/I4387153335"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yuxuan Wang","raw_affiliation_strings":["Institute of Artificial Intelligence (TeleAI), China Telecom, Shanghai, China","Shanghai Jiao Tong University, Shanghai, China","Zhiyuan College, Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0001-1595-411X","affiliations":[{"raw_affiliation_string":"Institute of Artificial Intelligence (TeleAI), China Telecom, Shanghai, China","institution_ids":["https://openalex.org/I4210136246","https://openalex.org/I4387153335"]},{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Zhiyuan College, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134505332","display_name":"Yanbo Wang","orcid":"https://orcid.org/0009-0002-9679-8602"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]},{"id":"https://openalex.org/I4210136246","display_name":"China Telecom (China)","ror":"https://ror.org/03jgnzt20","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210136246"]},{"id":"https://openalex.org/I4387153335","display_name":"China Telecom","ror":"https://ror.org/05p67dv18","country_code":null,"type":"company","lineage":["https://openalex.org/I4387153335"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanbo Wang","raw_affiliation_strings":["Institute of Artificial Intelligence (TeleAI), China Telecom, Shanghai, China","Shanghai Jiao Tong University, Shanghai, China","Zhiyuan College, Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0002-9679-8602","affiliations":[{"raw_affiliation_string":"Institute of Artificial Intelligence (TeleAI), China Telecom, Shanghai, China","institution_ids":["https://openalex.org/I4210136246","https://openalex.org/I4387153335"]},{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Zhiyuan College, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028452388","display_name":"C J Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen Chen","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0001-9480-5632","affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109042106","display_name":"C Y Xue","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chunyu Xue","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0008-9272-1732","affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002653936","display_name":"Qizhen Weng","orcid":"https://orcid.org/0000-0001-9195-6443"},"institutions":[{"id":"https://openalex.org/I4210136246","display_name":"China Telecom (China)","ror":"https://ror.org/03jgnzt20","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210136246"]},{"id":"https://openalex.org/I4387153335","display_name":"China Telecom","ror":"https://ror.org/05p67dv18","country_code":null,"type":"company","lineage":["https://openalex.org/I4387153335"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qizhen Weng","raw_affiliation_strings":["Institute of Artificial Intelligence (TeleAI), China Telecom, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0001-9195-6443","affiliations":[{"raw_affiliation_string":"Institute of Artificial Intelligence (TeleAI), China Telecom, Shanghai, China","institution_ids":["https://openalex.org/I4210136246","https://openalex.org/I4387153335"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046212412","display_name":"Yin Chen","orcid":"https://orcid.org/0000-0002-0367-4276"},"institutions":[{"id":"https://openalex.org/I4210136246","display_name":"China Telecom (China)","ror":"https://ror.org/03jgnzt20","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210136246"]},{"id":"https://openalex.org/I4387153335","display_name":"China Telecom","ror":"https://ror.org/05p67dv18","country_code":null,"type":"company","lineage":["https://openalex.org/I4387153335"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yin Chen","raw_affiliation_strings":["Institute of Artificial Intelligence (TeleAI), China Telecom, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0005-4930-1028","affiliations":[{"raw_affiliation_string":"Institute of Artificial Intelligence (TeleAI), China Telecom, Shanghai, China","institution_ids":["https://openalex.org/I4210136246","https://openalex.org/I4387153335"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134487862","display_name":"Zeren Li","orcid":"https://orcid.org/0009-0002-9287-9413"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zeren Li","raw_affiliation_strings":["Huawei Technologies Co., Ltd., Shenzhen, China"],"raw_orcid":"https://orcid.org/0009-0002-9287-9413","affiliations":[{"raw_affiliation_string":"Huawei Technologies Co., Ltd., Shenzhen, China","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101835782","display_name":"Xuqi Zhu","orcid":"https://orcid.org/0009-0007-1131-6237"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuqi Zhu","raw_affiliation_strings":["Huawei Technologies Co., Ltd., Shenzhen, China"],"raw_orcid":"https://orcid.org/0009-0007-1131-6237","affiliations":[{"raw_affiliation_string":"Huawei Technologies Co., Ltd., Shenzhen, China","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134487080","display_name":"Yongqiang Yang","orcid":"https://orcid.org/0000-0001-9733-4346"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yongqiang Yang","raw_affiliation_strings":["Huawei Technologies Co., Ltd., National Natural Science Foundation of Chinahen, China"],"raw_orcid":"https://orcid.org/0000-0001-9733-4346","affiliations":[{"raw_affiliation_string":"Huawei Technologies Co., Ltd., National Natural Science Foundation of Chinahen, China","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100377840","display_name":"Quan Chen","orcid":"https://orcid.org/0000-0001-5832-0347"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Quan Chen","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0001-5832-0347","affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5039318240","display_name":"Minyi Guo","orcid":"https://orcid.org/0000-0003-0034-2302"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Minyi Guo","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0003-0034-2302","affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":11,"corresponding_author_ids":["https://openalex.org/A5134544287"],"corresponding_institution_ids":["https://openalex.org/I183067930","https://openalex.org/I4210136246","https://openalex.org/I4387153335"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.94826101,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2002","last_page":"2021"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.5957000255584717,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.5957000255584717,"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":0.06629999727010727,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.03799999877810478,"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/training","display_name":"Training (meteorology)","score":0.5490000247955322},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.3142000138759613},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.30970001220703125},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.28929999470710754},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.2847000062465668},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.2685000002384186}],"concepts":[{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5490000247955322},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5478000044822693},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.3142000138759613},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.30970001220703125},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.2985999882221222},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.28929999470710754},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2847000062465668},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2685000002384186},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2680000066757202},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.26269999146461487},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.25920000672340393},{"id":"https://openalex.org/C195094911","wikidata":"https://www.wikidata.org/wiki/Q14167904","display_name":"Process management","level":1,"score":0.25540000200271606}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3767295.3803623","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767295.3803623","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st European Conference on Computer Systems","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3767295.3803623","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767295.3803623","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st European Conference on Computer Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.4799661338329315,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":5,"referenced_works":["https://openalex.org/W2150884987","https://openalex.org/W4384662964","https://openalex.org/W4391494845","https://openalex.org/W4403203873","https://openalex.org/W4415756055"],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"are":[4,21],"usually":[5],"trained":[6],"with":[7,94],"3D":[8],"(data,":[9],"tensor,":[10],"and":[11,51,63,104,115,145],"pipeline)":[12],"parallelism\u2014in":[13],"shared":[14],"GPU":[15],"clusters":[16],"where":[17],"the":[18,25,75,80,110,131],"available":[19],"resources":[20,27],"highly":[22],"dynamic.":[23],"Rescheduling":[24],"idle":[26],"to":[28,108,129,161],"ongoing":[29],"jobs":[30,42],"can":[31],"help":[32],"improve":[33],"cluster":[34,57,144,148],"utilization,":[35],"but":[36],"doing":[37],"so":[38],"for":[39,67,137],"3D-parallelized":[40,68],"training":[41,58,70],"suffers":[43],"large":[44],"overheads":[45],"in":[46,157],"performance":[47,86,92],"modeling,":[48,87],"decision":[49,120],"making,":[50],"redeployment.":[52],"We":[53],"present":[54],"Suika,":[55],"a":[56,125],"system":[59],"that":[60],"supports":[61],"efficient":[62],"high-quality":[64],"resource":[65,113],"rescheduling":[66],"LLM":[69],"jobs.":[71],"Suika":[72,122,151],"holistically":[73],"addresses":[74],"complexity":[76,111],"challenges":[77],"by":[78],"exploiting":[79],"incremental":[81,135],"nature":[82,133],"of":[83,112,134],"rescheduling.":[84],"For":[85,98],"it":[88,100],"builds":[89],"an":[90,105],"accurate":[91],"estimator":[93],"non-disruptive":[95],"online":[96],"profiling.":[97],"decision-making,":[99],"employs":[101],"topology-aware":[102],"sorting":[103],"expand-and-balance":[106],"algorithm":[107],"reduce":[109],"allocation":[114],"job":[116],"parallelization,":[117],"without":[118],"compromising":[119],"quality.":[121],"further":[123],"integrates":[124],"device-to-device":[126],"redeployment":[127],"method":[128],"leverage":[130],"overlapping":[132],"reconfiguration":[136],"overhead":[138],"reduction.":[139],"Experiments":[140],"on":[141],"64-GPU":[142],"physical":[143],"1024-GPU":[146],"simulated":[147],"show":[149],"that,":[150],"achieves":[152],"1.29":[153],"~":[154],"1.31\u00d7":[155],"reduction":[156],"average":[158],"JCT":[159],"compared":[160],"state-of-the-art":[162],"schedulers.":[163]},"counts_by_year":[],"updated_date":"2026-04-29T09:16:38.111599","created_date":"2026-04-25T00:00:00"}