{"id":"https://openalex.org/W3135674429","doi":"https://doi.org/10.14778/3476249.3476264","title":"Large graph convolutional network training with GPU-oriented data communication architecture","display_name":"Large graph convolutional network training with GPU-oriented data communication architecture","publication_year":2021,"publication_date":"2021-07-01","ids":{"openalex":"https://openalex.org/W3135674429","doi":"https://doi.org/10.14778/3476249.3476264","mag":"3135674429"},"language":"en","primary_location":{"id":"doi:10.14778/3476249.3476264","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3476249.3476264","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the VLDB Endowment","raw_type":"journal-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2103.03330","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5067853086","display_name":"Seung Won Min","orcid":null},"institutions":[{"id":"https://openalex.org/I308392441","display_name":"International University of the Caribbean","ror":"https://ror.org/02rv57d03","country_code":"JM","type":"education","lineage":["https://openalex.org/I308392441"]}],"countries":["JM"],"is_corresponding":true,"raw_author_name":"Seung Won Min","raw_affiliation_strings":["UIUC","UIUC,#TAB#"],"affiliations":[{"raw_affiliation_string":"UIUC","institution_ids":["https://openalex.org/I308392441"]},{"raw_affiliation_string":"UIUC,#TAB#","institution_ids":["https://openalex.org/I308392441"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073106039","display_name":"Kun Wu","orcid":"https://orcid.org/0000-0002-0149-1409"},"institutions":[{"id":"https://openalex.org/I308392441","display_name":"International University of the Caribbean","ror":"https://ror.org/02rv57d03","country_code":"JM","type":"education","lineage":["https://openalex.org/I308392441"]}],"countries":["JM"],"is_corresponding":false,"raw_author_name":"Kun Wu","raw_affiliation_strings":["UIUC","UIUC,#TAB#"],"affiliations":[{"raw_affiliation_string":"UIUC","institution_ids":["https://openalex.org/I308392441"]},{"raw_affiliation_string":"UIUC,#TAB#","institution_ids":["https://openalex.org/I308392441"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050532440","display_name":"Sitao Huang","orcid":"https://orcid.org/0000-0001-7669-1467"},"institutions":[{"id":"https://openalex.org/I308392441","display_name":"International University of the Caribbean","ror":"https://ror.org/02rv57d03","country_code":"JM","type":"education","lineage":["https://openalex.org/I308392441"]}],"countries":["JM"],"is_corresponding":false,"raw_author_name":"Sitao Huang","raw_affiliation_strings":["UIUC","UIUC,#TAB#"],"affiliations":[{"raw_affiliation_string":"UIUC","institution_ids":["https://openalex.org/I308392441"]},{"raw_affiliation_string":"UIUC,#TAB#","institution_ids":["https://openalex.org/I308392441"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024169920","display_name":"Mert Hidayeto\u011flu","orcid":"https://orcid.org/0000-0001-9276-5075"},"institutions":[{"id":"https://openalex.org/I308392441","display_name":"International University of the Caribbean","ror":"https://ror.org/02rv57d03","country_code":"JM","type":"education","lineage":["https://openalex.org/I308392441"]}],"countries":["JM"],"is_corresponding":false,"raw_author_name":"Mert Hidayeto\u011flu","raw_affiliation_strings":["UIUC","UIUC,#TAB#"],"affiliations":[{"raw_affiliation_string":"UIUC","institution_ids":["https://openalex.org/I308392441"]},{"raw_affiliation_string":"UIUC,#TAB#","institution_ids":["https://openalex.org/I308392441"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030156276","display_name":"Jinjun Xiong","orcid":"https://orcid.org/0000-0002-2620-4859"},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinjun Xiong","raw_affiliation_strings":["IBM T.J. Watson Research Center","IBM T.J. Watson Research Center, ,"],"affiliations":[{"raw_affiliation_string":"IBM T.J. Watson Research Center","institution_ids":[]},{"raw_affiliation_string":"IBM T.J. Watson Research Center, ,","institution_ids":["https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089789557","display_name":"Eiman Ebrahimi","orcid":null},"institutions":[{"id":"https://openalex.org/I1304085615","display_name":"Nvidia (United Kingdom)","ror":"https://ror.org/02kr42612","country_code":"GB","type":"company","lineage":["https://openalex.org/I1304085615","https://openalex.org/I4210127875"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Eiman Ebrahimi","raw_affiliation_strings":["NVIDIA","nVidia"],"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]},{"raw_affiliation_string":"nVidia","institution_ids":["https://openalex.org/I1304085615"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056321228","display_name":"Deming Chen","orcid":"https://orcid.org/0000-0002-3016-0270"},"institutions":[{"id":"https://openalex.org/I308392441","display_name":"International University of the Caribbean","ror":"https://ror.org/02rv57d03","country_code":"JM","type":"education","lineage":["https://openalex.org/I308392441"]}],"countries":["JM"],"is_corresponding":false,"raw_author_name":"Deming Chen","raw_affiliation_strings":["UIUC","UIUC,#TAB#"],"affiliations":[{"raw_affiliation_string":"UIUC","institution_ids":["https://openalex.org/I308392441"]},{"raw_affiliation_string":"UIUC,#TAB#","institution_ids":["https://openalex.org/I308392441"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5040404999","display_name":"Wen\u2010mei Hwu","orcid":"https://orcid.org/0000-0003-2532-5349"},"institutions":[{"id":"https://openalex.org/I308392441","display_name":"International University of the Caribbean","ror":"https://ror.org/02rv57d03","country_code":"JM","type":"education","lineage":["https://openalex.org/I308392441"]}],"countries":["JM"],"is_corresponding":false,"raw_author_name":"Wen-mei Hwu","raw_affiliation_strings":["UIUC","UIUC,#TAB#"],"affiliations":[{"raw_affiliation_string":"UIUC","institution_ids":["https://openalex.org/I308392441"]},{"raw_affiliation_string":"UIUC,#TAB#","institution_ids":["https://openalex.org/I308392441"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5067853086"],"corresponding_institution_ids":["https://openalex.org/I308392441"],"apc_list":null,"apc_paid":null,"fwci":0.8399,"has_fulltext":true,"cited_by_count":6,"citation_normalized_percentile":{"value":0.77843721,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":"14","issue":"11","first_page":"2087","last_page":"2100"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9995999932289124,"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"}},"topics":[{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9995999932289124,"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"}},{"id":"https://openalex.org/T11478","display_name":"Caching and Content Delivery","score":0.9969000220298767,"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/T10203","display_name":"Recommender Systems and Techniques","score":0.9943000078201294,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8958149552345276},{"id":"https://openalex.org/keywords/pci-express","display_name":"PCI Express","score":0.573157012462616},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5316192507743835},{"id":"https://openalex.org/keywords/uniform-memory-access","display_name":"Uniform memory access","score":0.47805482149124146},{"id":"https://openalex.org/keywords/central-processing-unit","display_name":"Central processing unit","score":0.41129326820373535},{"id":"https://openalex.org/keywords/memory-management","display_name":"Memory management","score":0.3111335039138794},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.29293495416641235},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.1718735694885254},{"id":"https://openalex.org/keywords/semiconductor-memory","display_name":"Semiconductor memory","score":0.15621009469032288}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8958149552345276},{"id":"https://openalex.org/C64270927","wikidata":"https://www.wikidata.org/wiki/Q206924","display_name":"PCI Express","level":3,"score":0.573157012462616},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5316192507743835},{"id":"https://openalex.org/C51290061","wikidata":"https://www.wikidata.org/wiki/Q1936765","display_name":"Uniform memory access","level":4,"score":0.47805482149124146},{"id":"https://openalex.org/C49154492","wikidata":"https://www.wikidata.org/wiki/Q5300","display_name":"Central processing unit","level":2,"score":0.41129326820373535},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.3111335039138794},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.29293495416641235},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.1718735694885254},{"id":"https://openalex.org/C98986596","wikidata":"https://www.wikidata.org/wiki/Q1143031","display_name":"Semiconductor memory","level":2,"score":0.15621009469032288}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.14778/3476249.3476264","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3476249.3476264","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the VLDB Endowment","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2103.03330","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2103.03330","pdf_url":"https://arxiv.org/pdf/2103.03330","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2103.03330","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2103.03330","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"},{"id":"mag:3135674429","is_oa":false,"landing_page_url":null,"pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":null}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2103.03330","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2103.03330","pdf_url":"https://arxiv.org/pdf/2103.03330","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3135674429.pdf","grobid_xml":"https://content.openalex.org/works/W3135674429.grobid-xml"},"referenced_works_count":47,"referenced_works":["https://openalex.org/W131619556","https://openalex.org/W1854214752","https://openalex.org/W2147800946","https://openalex.org/W2154851992","https://openalex.org/W2183341477","https://openalex.org/W2302255633","https://openalex.org/W2405578611","https://openalex.org/W2534888058","https://openalex.org/W2554952599","https://openalex.org/W2604319603","https://openalex.org/W2624431344","https://openalex.org/W2807021761","https://openalex.org/W2867345499","https://openalex.org/W2907492528","https://openalex.org/W2918342466","https://openalex.org/W2930869794","https://openalex.org/W2945827377","https://openalex.org/W2951136539","https://openalex.org/W2962756421","https://openalex.org/W2963460103","https://openalex.org/W2963581908","https://openalex.org/W2963695795","https://openalex.org/W2963803379","https://openalex.org/W2964015378","https://openalex.org/W2964145825","https://openalex.org/W2964174152","https://openalex.org/W2964311892","https://openalex.org/W2964321699","https://openalex.org/W2971196067","https://openalex.org/W2981587687","https://openalex.org/W2994968268","https://openalex.org/W3002924435","https://openalex.org/W3013163801","https://openalex.org/W3019011053","https://openalex.org/W3021182071","https://openalex.org/W3021975806","https://openalex.org/W3035128007","https://openalex.org/W3037699692","https://openalex.org/W3042369493","https://openalex.org/W3080555959","https://openalex.org/W3093166425","https://openalex.org/W3100848837","https://openalex.org/W3103894541","https://openalex.org/W3104097132","https://openalex.org/W3159953606","https://openalex.org/W4210257598","https://openalex.org/W4247353671"],"related_works":["https://openalex.org/W3035128007","https://openalex.org/W2908650493","https://openalex.org/W3012356635","https://openalex.org/W3040726395","https://openalex.org/W2931791443","https://openalex.org/W2276395270","https://openalex.org/W3090784541","https://openalex.org/W2050019858","https://openalex.org/W3158564519","https://openalex.org/W2782647461","https://openalex.org/W3112502950","https://openalex.org/W3030329349","https://openalex.org/W2785460572","https://openalex.org/W2755395451","https://openalex.org/W2999509507","https://openalex.org/W2954292268","https://openalex.org/W2033227022","https://openalex.org/W3091048326","https://openalex.org/W1988132189","https://openalex.org/W780385496"],"abstract_inverted_index":{"Graph":[0],"Convolutional":[1],"Networks":[2],"(GCNs)":[3],"are":[4],"increasingly":[5],"adopted":[6],"in":[7,49,133,259],"large-scale":[8],"graph-based":[9],"recommender":[10],"systems.":[11],"Training":[12],"GCN":[13,42,124],"requires":[14],"the":[15,22,36,46,55,65,78,83,105,110,145,153,156,177,240,249],"minibatch":[16],"generator":[17],"traversing":[18],"graphs":[19,33,217,256],"and":[20,52,77,102,159,190,195,211,225,245],"sampling":[21],"sparsely":[23],"located":[24],"neighboring":[25],"nodes":[26,224],"to":[27,57,64,86,109,169,185,198,221],"obtain":[28],"their":[29],"features.":[30],"Since":[31],"real-world":[32],"often":[34],"exceed":[35],"capacity":[37],"of":[38,155,251],"GPU":[39,127,260],"memory,":[40,92],"current":[41],"training":[43,232,253],"systems":[44],"keep":[45],"feature":[47],"table":[48],"host":[50,74,134,157,172],"memory":[51,75,97,108,135,173],"rely":[53],"on":[54,73],"CPU":[56,84,141,146],"collect":[58],"sparse":[59,89,131],"features":[60,90,95,106,132],"before":[61],"sending":[62],"them":[63],"GPUs.":[66,111],"This":[67,80],"approach,":[68],"however,":[69],"puts":[70],"tremendous":[71],"pressure":[72],"bandwidth":[76],"CPU.":[79],"is":[81,236],"because":[82],"needs":[85],"(1)":[87,179],"read":[88],"from":[91,107],"(2)":[93,191],"write":[94],"into":[96,209],"as":[98],"a":[99,117,230],"dense":[100],"format,":[101],"(3)":[103],"transfer":[104,202,243],"In":[112,229],"this":[113],"work,":[114],"we":[115],"propose":[116],"novel":[118],"GPU-oriented":[119],"data":[120,160,181,201,242],"communication":[121],"approach":[122],"for":[123,254],"training,":[125],"where":[126],"threads":[128],"directly":[129],"access":[130,161,174,182,194],"through":[136],"zero-copy":[137,193],"accesses":[138],"without":[139],"much":[140],"help.":[142],"By":[143],"removing":[144],"gathering":[147],"stage,":[148],"our":[149,207,234],"method":[150,208,235],"significantly":[151],"reduces":[152],"consumption":[154],"resources":[158],"latency.":[162],"We":[163,205],"further":[164],"present":[165],"two":[166],"important":[167],"techniques":[168],"achieve":[170],"high":[171],"efficiency":[175],"by":[176],"GPU:":[178],"automatic":[180],"address":[183],"alignment":[184],"maximize":[186],"PCIe":[187],"packet":[188],"efficiency,":[189],"asynchronous":[192],"kernel":[196],"execution":[197],"fully":[199],"overlap":[200],"with":[203,218],"training.":[204],"incorporate":[206],"PyTorch":[210],"evaluate":[212],"its":[213],"effectiveness":[214],"using":[215],"several":[216],"sizes":[219],"up":[220],"111":[222],"million":[223],"1.6":[226],"billion":[227],"edges.":[228],"multi-GPU":[231],"setup,":[233],"65--92%":[237],"faster":[238],"than":[239],"conventional":[241],"method,":[244],"can":[246],"even":[247],"match":[248],"performance":[250],"all-in-GPU-memory":[252],"some":[255],"that":[257],"fit":[258],"memory.":[261]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":2}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-10-10T00:00:00"}