{"id":"https://openalex.org/W4360831977","doi":"https://doi.org/10.1109/hpca56546.2023.10071043","title":"Tensor Movement Orchestration in Multi-GPU Training Systems","display_name":"Tensor Movement Orchestration in Multi-GPU Training Systems","publication_year":2023,"publication_date":"2023-02-01","ids":{"openalex":"https://openalex.org/W4360831977","doi":"https://doi.org/10.1109/hpca56546.2023.10071043"},"language":"en","primary_location":{"id":"doi:10.1109/hpca56546.2023.10071043","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca56546.2023.10071043","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","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/A5036375505","display_name":"Shaofu Lin","orcid":"https://orcid.org/0000-0002-3653-6531"},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":true,"raw_author_name":"Shao-Fu Lin","raw_affiliation_strings":["National Taiwan University,Department of Computer Science and Information Engineering,Taipei,Taiwan","Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan"],"affiliations":[{"raw_affiliation_string":"National Taiwan University,Department of Computer Science and Information Engineering,Taipei,Taiwan","institution_ids":["https://openalex.org/I16733864"]},{"raw_affiliation_string":"Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114859256","display_name":"Yi-Jung Chen","orcid":"https://orcid.org/0000-0002-3646-7716"},"institutions":[{"id":"https://openalex.org/I193365169","display_name":"National Chi Nan University","ror":"https://ror.org/03ha6v181","country_code":"TW","type":"education","lineage":["https://openalex.org/I193365169"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Yi-Jung Chen","raw_affiliation_strings":["National Chi Nan University,Department of Computer Science and Information Engineering,Nantou,Taiwan","Department of Computer Science and Information Engineering, National Chi Nan University, Nantou, Taiwan"],"affiliations":[{"raw_affiliation_string":"National Chi Nan University,Department of Computer Science and Information Engineering,Nantou,Taiwan","institution_ids":["https://openalex.org/I193365169"]},{"raw_affiliation_string":"Department of Computer Science and Information Engineering, National Chi Nan University, Nantou, Taiwan","institution_ids":["https://openalex.org/I193365169"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043930707","display_name":"Hsiang-Yun Cheng","orcid":"https://orcid.org/0000-0002-8983-9835"},"institutions":[{"id":"https://openalex.org/I4210086894","display_name":"Research Center for Information Technology Innovation, Academia Sinica","ror":"https://ror.org/000zgvm20","country_code":"TW","type":"facility","lineage":["https://openalex.org/I4210086894","https://openalex.org/I84653119"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Hsiang-Yun Cheng","raw_affiliation_strings":["Academia Sinica,Research Center for Information Technology Innovation,Taipei,Taiwan","Research Center for Information Technology Innovation, Academia Sinica, Taipei, Taiwan"],"affiliations":[{"raw_affiliation_string":"Academia Sinica,Research Center for Information Technology Innovation,Taipei,Taiwan","institution_ids":["https://openalex.org/I4210086894"]},{"raw_affiliation_string":"Research Center for Information Technology Innovation, Academia Sinica, Taipei, Taiwan","institution_ids":["https://openalex.org/I4210086894"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5046948316","display_name":"Chia-Lin Yang","orcid":"https://orcid.org/0000-0003-0091-5027"},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Chia-Lin Yang","raw_affiliation_strings":["National Taiwan University,Department of Computer Science and Information Engineering,Taipei,Taiwan","Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan"],"affiliations":[{"raw_affiliation_string":"National Taiwan University,Department of Computer Science and Information Engineering,Taipei,Taiwan","institution_ids":["https://openalex.org/I16733864"]},{"raw_affiliation_string":"Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan","institution_ids":["https://openalex.org/I16733864"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5036375505"],"corresponding_institution_ids":["https://openalex.org/I16733864"],"apc_list":null,"apc_paid":null,"fwci":1.2418,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.7600921,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1140","last_page":"1152"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9987000226974487,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9987000226974487,"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/T12303","display_name":"Tensor decomposition and applications","score":0.9984999895095825,"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/T11304","display_name":"Advanced Neuroimaging Techniques and Applications","score":0.9855999946594238,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8688918352127075},{"id":"https://openalex.org/keywords/pci-express","display_name":"PCI Express","score":0.786818265914917},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6131939888000488},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.5248832106590271},{"id":"https://openalex.org/keywords/disjoint-sets","display_name":"Disjoint sets","score":0.476857990026474},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.4239703416824341},{"id":"https://openalex.org/keywords/host","display_name":"Host (biology)","score":0.4214687943458557},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.28310543298721313},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.2777688801288605},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.12210345268249512}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8688918352127075},{"id":"https://openalex.org/C64270927","wikidata":"https://www.wikidata.org/wiki/Q206924","display_name":"PCI Express","level":3,"score":0.786818265914917},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6131939888000488},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.5248832106590271},{"id":"https://openalex.org/C45340560","wikidata":"https://www.wikidata.org/wiki/Q215382","display_name":"Disjoint sets","level":2,"score":0.476857990026474},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.4239703416824341},{"id":"https://openalex.org/C126831891","wikidata":"https://www.wikidata.org/wiki/Q221673","display_name":"Host (biology)","level":2,"score":0.4214687943458557},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.28310543298721313},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.2777688801288605},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.12210345268249512},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C18903297","wikidata":"https://www.wikidata.org/wiki/Q7150","display_name":"Ecology","level":1,"score":0.0},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpca56546.2023.10071043","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca56546.2023.10071043","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320307110","display_name":"Delta","ror":"https://ror.org/03g9c1e75"},{"id":"https://openalex.org/F4320308737","display_name":"Facebook","ror":"https://ror.org/01zbnvs85"},{"id":"https://openalex.org/F4320309618","display_name":"Ministry of Science and Technology","ror":"https://ror.org/02b207r52"},{"id":"https://openalex.org/F4320311649","display_name":"Ministry of Education","ror":"https://ror.org/036nq5137"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":50,"referenced_works":["https://openalex.org/W1598866093","https://openalex.org/W1988622494","https://openalex.org/W2194775991","https://openalex.org/W2336650964","https://openalex.org/W2338908902","https://openalex.org/W2339765813","https://openalex.org/W2489529491","https://openalex.org/W2753239506","https://openalex.org/W2883283076","https://openalex.org/W2896457183","https://openalex.org/W2920536219","https://openalex.org/W2947737663","https://openalex.org/W2951882630","https://openalex.org/W2963446712","https://openalex.org/W2965373594","https://openalex.org/W2969388332","https://openalex.org/W2973727699","https://openalex.org/W2991040477","https://openalex.org/W3010830594","https://openalex.org/W3012514909","https://openalex.org/W3036879053","https://openalex.org/W3086105743","https://openalex.org/W3094502228","https://openalex.org/W3097777922","https://openalex.org/W3107819383","https://openalex.org/W3129831491","https://openalex.org/W3157864729","https://openalex.org/W3205803342","https://openalex.org/W4289646389","https://openalex.org/W4295312788","https://openalex.org/W4301361180","https://openalex.org/W4312443924","https://openalex.org/W6635810480","https://openalex.org/W6703420464","https://openalex.org/W6703652217","https://openalex.org/W6713134421","https://openalex.org/W6755207826","https://openalex.org/W6756379755","https://openalex.org/W6756718674","https://openalex.org/W6760340630","https://openalex.org/W6763737044","https://openalex.org/W6766673545","https://openalex.org/W6766978945","https://openalex.org/W6767997687","https://openalex.org/W6778883912","https://openalex.org/W6779863968","https://openalex.org/W6780218876","https://openalex.org/W6784333009","https://openalex.org/W6787953186","https://openalex.org/W6862640317"],"related_works":["https://openalex.org/W4385894176","https://openalex.org/W2347371119","https://openalex.org/W2612768808","https://openalex.org/W3131402800","https://openalex.org/W2388965158","https://openalex.org/W2808480228","https://openalex.org/W2366201982","https://openalex.org/W2593071546","https://openalex.org/W2805641962","https://openalex.org/W2012850936"],"abstract_inverted_index":{"As":[0],"deep":[1],"neural":[2,18],"network":[3],"(DNN)":[4],"models":[5],"grow":[6],"deeper":[7],"and":[8,43,143],"wider,":[9],"one":[10],"of":[11,45,51,98,134,171],"the":[12,32,36,49,66,71,78,102,122,140,146,165,176,185],"main":[13],"challenges":[14],"for":[15,39,118],"training":[16,60],"large-scale":[17],"networks":[19],"is":[20,29,160,180],"overcoming":[21],"limited":[22],"GPU":[23,46,186],"memory":[24,34,38,177,187],"capacity.":[25],"One":[26],"common":[27],"solution":[28,167],"to":[30,63,70,89,129,156,162,173,182],"utilize":[31],"host":[33],"as":[35],"external":[37],"swapping":[40,54],"tensors":[41,117],"in":[42,58],"out":[44],"memory.":[47],"However,":[48],"effectiveness":[50,133],"such":[52],"tensor":[53,85],"can":[55],"be":[56],"impaired":[57],"data-parallel":[59],"systems":[61],"due":[62],"contention":[64],"on":[65,139,149],"shared":[67],"PCIe":[68,91,105],"channel":[69,92,106,163],"host.":[72],"In":[73,101,121],"this":[74],"paper,":[75],"we":[76],"propose":[77],"first":[79,103],"large-model":[80,157],"support":[81,158],"framework":[82],"that":[83,154,159],"coordinates":[84],"movements":[86],"among":[87],"GPUs":[88,110,147],"alleviate":[90],"contention.":[93,131],"We":[94],"design":[95],"two":[96,136],"types":[97],"coordination":[99],"mechanisms.":[100],"mechanism,":[104],"accesses":[107],"from":[108],"different":[109],"are":[111,127],"interleaved":[112],"by":[113],"selecting":[114],"disjoint":[115],"swapped-out":[116],"each":[119],"GPU.":[120],"second":[123],"method,":[124],"swap":[125],"commands":[126],"orchestrated":[128],"avoid":[130],"The":[132],"these":[135],"methods":[137],"depends":[138],"model":[141],"size":[142,179],"how":[144],"often":[145],"synchronize":[148],"gradients.":[150],"Experimental":[151],"results":[152],"show":[153],"compared":[155],"oblivious":[161],"contention,":[164],"proposed":[166],"achieves":[168],"average":[169],"speedups":[170],"38.3%":[172],"31.8%":[174],"when":[175],"footprint":[178],"1.33":[181],"2":[183],"times":[184],"size.":[188]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3}],"updated_date":"2026-03-06T13:50:29.536080","created_date":"2025-10-10T00:00:00"}
