{"id":"https://openalex.org/W4290944486","doi":"https://doi.org/10.1145/3534678.3539177","title":"Distributed Hybrid CPU and GPU training for Graph Neural Networks on Billion-Scale Heterogeneous Graphs","display_name":"Distributed Hybrid CPU and GPU training for Graph Neural Networks on Billion-Scale Heterogeneous Graphs","publication_year":2022,"publication_date":"2022-08-12","ids":{"openalex":"https://openalex.org/W4290944486","doi":"https://doi.org/10.1145/3534678.3539177"},"language":"en","primary_location":{"id":"doi:10.1145/3534678.3539177","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3534678.3539177","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3534678.3539177","source":{"id":"https://openalex.org/S4363608767","display_name":"Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3534678.3539177","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5060763203","display_name":"Da Zheng","orcid":"https://orcid.org/0000-0001-8115-5415"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Da Zheng","raw_affiliation_strings":["Amazon, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"Amazon, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064457872","display_name":"Xiang Song","orcid":"https://orcid.org/0000-0002-1704-4339"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiang Song","raw_affiliation_strings":["Amazon, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"Amazon, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000734067","display_name":"Chengru Yang","orcid":"https://orcid.org/0009-0003-4320-0723"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chengru Yang","raw_affiliation_strings":["Amazon, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Amazon, Shanghai, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053618309","display_name":"Dominique LaSalle","orcid":null},"institutions":[{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dominique LaSalle","raw_affiliation_strings":["NVIDIA Corporation, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"NVIDIA Corporation, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5082384108","display_name":"George Karypis","orcid":"https://orcid.org/0000-0003-2753-1437"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"George Karypis","raw_affiliation_strings":["Amazon, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"Amazon, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I1311688040"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5060763203"],"corresponding_institution_ids":["https://openalex.org/I1311688040"],"apc_list":null,"apc_paid":null,"fwci":3.2256,"has_fulltext":true,"cited_by_count":62,"citation_normalized_percentile":{"value":0.94272534,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"4582","last_page":"4591"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12292","display_name":"Graph Theory and Algorithms","score":0.9994999766349792,"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"}},"topics":[{"id":"https://openalex.org/T12292","display_name":"Graph Theory and Algorithms","score":0.9994999766349792,"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"}},{"id":"https://openalex.org/T11522","display_name":"VLSI and FPGA Design Techniques","score":0.9987999796867371,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9986000061035156,"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/computer-science","display_name":"Computer science","score":0.8587143421173096},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6660764813423157},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5816085338592529},{"id":"https://openalex.org/keywords/asynchronous-communication","display_name":"Asynchronous communication","score":0.5502054691314697},{"id":"https://openalex.org/keywords/load-balancing","display_name":"Load balancing (electrical power)","score":0.5350099205970764},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.47858190536499023},{"id":"https://openalex.org/keywords/pci-express","display_name":"PCI Express","score":0.44069281220436096},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.35358643531799316},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.20946887135505676},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.19789880514144897},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.16229209303855896}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8587143421173096},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6660764813423157},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5816085338592529},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.5502054691314697},{"id":"https://openalex.org/C138959212","wikidata":"https://www.wikidata.org/wiki/Q1806783","display_name":"Load balancing (electrical power)","level":3,"score":0.5350099205970764},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.47858190536499023},{"id":"https://openalex.org/C64270927","wikidata":"https://www.wikidata.org/wiki/Q206924","display_name":"PCI Express","level":3,"score":0.44069281220436096},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.35358643531799316},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.20946887135505676},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.19789880514144897},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.16229209303855896},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C187691185","wikidata":"https://www.wikidata.org/wiki/Q2020720","display_name":"Grid","level":2,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3534678.3539177","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3534678.3539177","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3534678.3539177","source":{"id":"https://openalex.org/S4363608767","display_name":"Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3534678.3539177","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3534678.3539177","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3534678.3539177","source":{"id":"https://openalex.org/S4363608767","display_name":"Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.6200000047683716}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4290944486.pdf","grobid_xml":"https://content.openalex.org/works/W4290944486.grobid-xml"},"referenced_works_count":16,"referenced_works":["https://openalex.org/W1583837637","https://openalex.org/W2070232376","https://openalex.org/W2127391575","https://openalex.org/W2154987621","https://openalex.org/W2604314403","https://openalex.org/W2734941459","https://openalex.org/W2951136539","https://openalex.org/W3037699692","https://openalex.org/W3096566397","https://openalex.org/W3164865299","https://openalex.org/W3206504463","https://openalex.org/W4205241946","https://openalex.org/W6736685754","https://openalex.org/W6745537798","https://openalex.org/W6762631216","https://openalex.org/W6796909790"],"related_works":["https://openalex.org/W4246549241","https://openalex.org/W2168758875","https://openalex.org/W2410733619","https://openalex.org/W2963483475","https://openalex.org/W2317245370","https://openalex.org/W4249323025","https://openalex.org/W198851386","https://openalex.org/W2030310580","https://openalex.org/W947442053","https://openalex.org/W1980160788"],"abstract_inverted_index":{"Graph":[0,102],"neural":[1],"networks":[2],"(GNN)":[3],"have":[4],"shown":[5],"great":[6],"success":[7],"in":[8,18,69,82,90],"learn-":[9],"ing":[10],"from":[11],"graph-structured":[12],"data.":[13],"They":[14],"are":[15,33],"widely":[16],"used":[17],"various":[19,191],"applications,":[20],"such":[21],"as":[22],"recommendation,":[23],"fraud":[24],"detection,":[25],"and":[26,36,45,86,107,130,143,159,174,206],"search.":[27],"In":[28],"these":[29],"domains,":[30],"the":[31],"graphs":[32,68,149,223],"typically":[34],"large":[35],"heterogeneous,":[37],"containing":[38],"many":[39],"millions":[40,227],"or":[41],"billions":[42],"of":[43,47,94,226,228],"vertices":[44,229],"edges":[46],"different":[48],"types.":[49],"To":[50,120,139],"tackle":[51],"this":[52],"challenge,":[53],"we":[54],"develop":[55],"DistDGLv2,":[56],"a":[57,70,126,152,231],"system":[58],"that":[59,171,197],"extends":[60],"DistDGL":[61,205],"for":[62],"training":[63,106,114,128],"GNNs":[64],"on":[65,190,222,230],"massive":[66],"heterogeneous":[67,108,148],"mini-batch":[71,88,105],"fashion,":[72],"using":[73,151],"distributed":[74,83,113],"hybrid":[75],"CPU/GPU":[76],"training.":[77],"DistDGLv2":[78,96,124,146,163,189,198],"places":[79],"graph":[80,109],"data":[81,141,175],"CPU":[84],"memory":[85],"performs":[87],"computation":[89,173],"GPUs.":[91,235],"For":[92],"ease":[93],"use,":[95],"adopts":[97],"API":[98],"compatible":[99],"with":[100,115,156,224,233],"Deep":[101],"Library":[103],"(DGL)'s":[104],"API,":[110],"which":[111],"enables":[112],"almost":[116],"no":[117],"code":[118],"modification.":[119],"ensure":[121,140],"model":[122],"accuracy,":[123],"follows":[125],"synchronous":[127],"approach":[129],"allows":[131],"ego-networks":[132],"forming":[133],"mini-batches":[134],"to":[135,178,218],"include":[136],"non-local":[137],"vertices.":[138],"locality":[142],"load":[144],"balancing,":[145],"partitions":[147],"by":[150],"multi-level":[153],"partitioning":[154],"algorithm":[155],"min-edge":[157],"cut":[158],"multiple":[160],"balancing":[161],"constraints.":[162],"deploys":[164],"an":[165,220],"asynchronous":[166,177],"mini-":[167],"batch":[168],"generation":[169],"pipeline":[170],"makes":[172],"access":[176],"fully":[179],"utilize":[180],"all":[181],"hardware":[182],"(CPU,":[183],"GPU,":[184],"network,":[185],"PCIe).":[186],"We":[187],"demonstrate":[188],"GNN":[192],"workloads.":[193],"Our":[194],"results":[195],"show":[196],"achieves":[199],"2":[200],"-":[201,215],"3x":[202],"speedup":[203,208],"over":[204,209],"18\u00d7":[207],"Euler.":[210],"It":[211],"takes":[212],"only":[213],"5":[214],"10":[216],"seconds":[217],"complete":[219],"epoch":[221],"hundreds":[225],"cluster":[232],"64":[234]},"counts_by_year":[{"year":2026,"cited_by_count":8},{"year":2025,"cited_by_count":24},{"year":2024,"cited_by_count":19},{"year":2023,"cited_by_count":8},{"year":2022,"cited_by_count":3}],"updated_date":"2026-03-28T08:17:26.163206","created_date":"2025-10-10T00:00:00"}