{"id":"https://openalex.org/W7125579285","doi":"https://doi.org/10.1109/tbdata.2026.3657302","title":"Accelerating Heterogeneous Tensor Parallelism via Flexible Workload Control","display_name":"Accelerating Heterogeneous Tensor Parallelism via Flexible Workload Control","publication_year":2026,"publication_date":"2026-01-23","ids":{"openalex":"https://openalex.org/W7125579285","doi":"https://doi.org/10.1109/tbdata.2026.3657302"},"language":null,"primary_location":{"id":"doi:10.1109/tbdata.2026.3657302","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tbdata.2026.3657302","pdf_url":null,"source":{"id":"https://openalex.org/S2491400915","display_name":"IEEE Transactions on Big Data","issn_l":"2332-7790","issn":["2332-7790","2372-2096"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","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 Big Data","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":null,"display_name":"Zhigang Wang","orcid":"https://orcid.org/0000-0002-1812-1068"},"institutions":[{"id":"https://openalex.org/I37987034","display_name":"Guangzhou University","ror":"https://ror.org/05ar8rn06","country_code":"CN","type":"education","lineage":["https://openalex.org/I37987034"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhigang Wang","raw_affiliation_strings":["Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-1812-1068","affiliations":[{"raw_affiliation_string":"Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China","institution_ids":["https://openalex.org/I37987034"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100437261","display_name":"Xu Zhang","orcid":"https://orcid.org/0000-0002-8098-0410"},"institutions":[{"id":"https://openalex.org/I59028903","display_name":"Ocean University of China","ror":"https://ror.org/04rdtx186","country_code":"CN","type":"education","lineage":["https://openalex.org/I59028903"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu Zhang","raw_affiliation_strings":["Faculty of Information Science and Engineering, Ocean University of China, Qingdao, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Faculty of Information Science and Engineering, Ocean University of China, Qingdao, China","institution_ids":["https://openalex.org/I59028903"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Ning Wang","orcid":"https://orcid.org/0000-0001-9698-7310"},"institutions":[{"id":"https://openalex.org/I37987034","display_name":"Guangzhou University","ror":"https://ror.org/05ar8rn06","country_code":"CN","type":"education","lineage":["https://openalex.org/I37987034"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ning Wang","raw_affiliation_strings":["Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China"],"raw_orcid":"https://orcid.org/0000-0001-9698-7310","affiliations":[{"raw_affiliation_string":"Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China","institution_ids":["https://openalex.org/I37987034"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Chuanfei Xu","orcid":"https://orcid.org/0000-0003-2653-0387"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chuanfei Xu","raw_affiliation_strings":["Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ), Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0003-2653-0387","affiliations":[{"raw_affiliation_string":"Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ), Shenzhen, China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123756378","display_name":"Yu Gu","orcid":null},"institutions":[{"id":"https://openalex.org/I9224756","display_name":"Northeastern University","ror":"https://ror.org/03awzbc87","country_code":"CN","type":"education","lineage":["https://openalex.org/I9224756"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu Gu","raw_affiliation_strings":["School of Computer Science and Engineering, Northeastern University, Shenyang, China"],"raw_orcid":"https://orcid.org/0000-0001-7422-6254","affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Northeastern University, Shenyang, China","institution_ids":["https://openalex.org/I9224756"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123752356","display_name":"Hui Lu","orcid":null},"institutions":[{"id":"https://openalex.org/I37987034","display_name":"Guangzhou University","ror":"https://ror.org/05ar8rn06","country_code":"CN","type":"education","lineage":["https://openalex.org/I37987034"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hui Lu","raw_affiliation_strings":["Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-4120-7716","affiliations":[{"raw_affiliation_string":"Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China","institution_ids":["https://openalex.org/I37987034"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123742249","display_name":"Dawei Zhao","orcid":null},"institutions":[{"id":"https://openalex.org/I152269853","display_name":"Qilu University of Technology","ror":"https://ror.org/04hyzq608","country_code":"CN","type":"education","lineage":["https://openalex.org/I152269853"]},{"id":"https://openalex.org/I4210142748","display_name":"Shandong Academy of Sciences","ror":"https://ror.org/04y8d6y55","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210142748"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dawei Zhao","raw_affiliation_strings":["Key Laboratory of Computing Power Network and Information Security, Ministry of Education, Qilu University of Technology (Shandong Academy of Sciences), Jinan, China"],"raw_orcid":"https://orcid.org/0000-0002-1812-1316","affiliations":[{"raw_affiliation_string":"Key Laboratory of Computing Power Network and Information Security, Ministry of Education, Qilu University of Technology (Shandong Academy of Sciences), Jinan, China","institution_ids":["https://openalex.org/I152269853","https://openalex.org/I4210142748"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5121836156","display_name":"Zhihong Tian","orcid":null},"institutions":[{"id":"https://openalex.org/I37987034","display_name":"Guangzhou University","ror":"https://ror.org/05ar8rn06","country_code":"CN","type":"education","lineage":["https://openalex.org/I37987034"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhihong Tian","raw_affiliation_strings":["Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-9409-5359","affiliations":[{"raw_affiliation_string":"Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China","institution_ids":["https://openalex.org/I37987034"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.125,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"12","issue":"3","first_page":"994","last_page":"1005"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12303","display_name":"Tensor decomposition and applications","score":0.3328000009059906,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12303","display_name":"Tensor decomposition and applications","score":0.3328000009059906,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.22709999978542328,"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"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.19539999961853027,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/workload","display_name":"Workload","score":0.7382000088691711},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.6682999730110168},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.6324999928474426},{"id":"https://openalex.org/keywords/parallelism","display_name":"Parallelism (grammar)","score":0.5685999989509583},{"id":"https://openalex.org/keywords/dependency","display_name":"Dependency (UML)","score":0.4810999929904938},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.47360000014305115},{"id":"https://openalex.org/keywords/resource","display_name":"Resource (disambiguation)","score":0.43700000643730164},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.36149999499320984},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.3531000018119812}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8950999975204468},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.7382000088691711},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6682999730110168},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.6324999928474426},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5927000045776367},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5753999948501587},{"id":"https://openalex.org/C2781172179","wikidata":"https://www.wikidata.org/wiki/Q853109","display_name":"Parallelism (grammar)","level":2,"score":0.5685999989509583},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.4810999929904938},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.47360000014305115},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.43700000643730164},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.36149999499320984},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.3531000018119812},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.351500004529953},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.35089999437332153},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.34290000796318054},{"id":"https://openalex.org/C106515295","wikidata":"https://www.wikidata.org/wiki/Q26806595","display_name":"Parallel processing","level":2,"score":0.3375999927520752},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3237000107765198},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3206999897956848},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3163999915122986},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.31439998745918274},{"id":"https://openalex.org/C2780870223","wikidata":"https://www.wikidata.org/wiki/Q1004415","display_name":"Runtime system","level":2,"score":0.3052999973297119},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.30469998717308044},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.29490000009536743},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.2720000147819519},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.27140000462532043},{"id":"https://openalex.org/C2781057727","wikidata":"https://www.wikidata.org/wiki/Q9941","display_name":"Parallel I/O","level":2,"score":0.2711000144481659},{"id":"https://openalex.org/C2780609101","wikidata":"https://www.wikidata.org/wiki/Q17156588","display_name":"Resource management (computing)","level":2,"score":0.267300009727478},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.26339998841285706},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.2606000006198883},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.26010000705718994},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.25270000100135803},{"id":"https://openalex.org/C2984822820","wikidata":"https://www.wikidata.org/wiki/Q1123036","display_name":"Processor scheduling","level":3,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tbdata.2026.3657302","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tbdata.2026.3657302","pdf_url":null,"source":{"id":"https://openalex.org/S2491400915","display_name":"IEEE Transactions on Big Data","issn_l":"2332-7790","issn":["2332-7790","2372-2096"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","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 Big Data","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.40981391072273254}],"awards":[{"id":"https://openalex.org/G1416658616","display_name":null,"funder_award_id":"62272121","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G1502308759","display_name":null,"funder_award_id":"U25A20424","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3791426778","display_name":null,"funder_award_id":"62572108","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6064493666","display_name":null,"funder_award_id":"U2436208","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G878108176","display_name":null,"funder_award_id":"62572139","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8812891240","display_name":null,"funder_award_id":"62372129","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W2019380637","https://openalex.org/W2083842231","https://openalex.org/W2523435939","https://openalex.org/W2584979567","https://openalex.org/W2768378004","https://openalex.org/W2888482885","https://openalex.org/W2902073008","https://openalex.org/W2963495533","https://openalex.org/W2979230481","https://openalex.org/W2991040477","https://openalex.org/W3009014747","https://openalex.org/W3043515497","https://openalex.org/W3109982791","https://openalex.org/W3116992856","https://openalex.org/W3124226966","https://openalex.org/W4220716191","https://openalex.org/W4294982491","https://openalex.org/W4312910503","https://openalex.org/W4313417875","https://openalex.org/W4319068657","https://openalex.org/W4321192305","https://openalex.org/W4378979503","https://openalex.org/W4386709668","https://openalex.org/W4386902904","https://openalex.org/W4394896940","https://openalex.org/W4395020669","https://openalex.org/W4399574191","https://openalex.org/W4405103169","https://openalex.org/W7133212984"],"related_works":[],"abstract_inverted_index":{"Transformer-based":[0],"foundation":[1],"models":[2],"are":[3,16,143],"becoming":[4],"deeper":[5],"and":[6,97,115,145,167,175],"larger.":[7],"For":[8],"fast":[9],"training,":[10],"their":[11],"billions":[12],"of":[13,109,136,165,189],"parameters":[14,110],"(tensors)":[15],"split":[17,75],"onto":[18],"parallel":[19],"tasks":[20],"running":[21],"on":[22,83,111,163],"many":[23],"modern":[24],"yet":[25],"expensive":[26],"accelerators.":[27],"To":[28],"amortize":[29],"the":[30,39,48,57,66,72,80,93,98,154,171,181,187],"huge":[31],"hardware":[32],"investments,":[33],"it":[34,177],"is":[35,79,105,122,161],"cost-effective":[36],"to":[37,71,106,113,152],"share":[38],"aggregated":[40],"resources":[41],"among":[42,74],"multi-tenants.":[43],"However,":[44],"resource":[45],"contention":[46],"yields":[47],"heavy":[49,99],"straggling":[50],"problem.":[51],"Existing":[52],"works":[53],"feature":[54],"contributions":[55],"for":[56,65],"traditional":[58],"data":[59],"parallelism.":[60,87],"They":[61],"cannot":[62],"work":[63],"well":[64],"new":[67],"tensor":[68,86,100],"parallelism,":[69],"due":[70],"dependency":[73],"tensors.":[76],"This":[77],"paper":[78],"first":[81],"attempt":[82],"accelerating":[84],"heterogeneous":[85],"We":[88,127],"summarize":[89],"specific":[90],"challenges,":[91],"including":[92],"very":[94],"frequent":[95],"synchronizations":[96],"computation":[101],"workloads.":[102,118],"Our":[103,157],"solution":[104,160],"resize":[107],"dimensions":[108],"demand,":[112],"quickly":[114],"dynamically":[116],"balance":[117],"The":[119,138],"accuracy":[120,174],"loss":[121,135],"reduced":[123],"through":[124],"priority":[125],"resizing.":[126],"also":[128],"migrate":[129],"workloads":[130],"between":[131,173],"tasks,":[132],"without":[133],"any":[134],"accuracy.":[137],"most":[139],"efficient":[140],"communication":[141],"primitives":[142],"selected":[144],"then":[146],"scheduled":[147],"in":[148],"a":[149],"non-redundant":[150],"manner,":[151],"reduce":[153],"runtime":[155],"latency.":[156],"final":[158],"hybrid":[159],"built":[162],"top":[164],"resizing":[166],"migration.":[168],"By":[169],"studying":[170],"tradeoff":[172],"efficiency,":[176],"can":[178],"smartly":[179],"hit":[180],"\u201csweet":[182],"spot\u201d.":[183],"Extensive":[184],"experiments":[185],"validate":[186],"effectiveness":[188],"our":[190],"proposals.":[191]},"counts_by_year":[],"updated_date":"2026-05-15T06:05:50.897203","created_date":"2026-01-25T00:00:00"}