{"id":"https://openalex.org/W7133533127","doi":"https://doi.org/10.1109/hpca68181.2026.11408524","title":"VeloxGNN: Efficient Out-of-Core GNN Training with Delayed Gradient Propagation","display_name":"VeloxGNN: Efficient Out-of-Core GNN Training with Delayed Gradient Propagation","publication_year":2026,"publication_date":"2026-01-31","ids":{"openalex":"https://openalex.org/W7133533127","doi":"https://doi.org/10.1109/hpca68181.2026.11408524"},"language":null,"primary_location":{"id":"doi:10.1109/hpca68181.2026.11408524","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408524","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 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/A5128126923","display_name":"Yi Li","orcid":null},"institutions":[{"id":"https://openalex.org/I162577319","display_name":"The University of Texas at Dallas","ror":"https://ror.org/049emcs32","country_code":"US","type":"education","lineage":["https://openalex.org/I162577319"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Yi Li","raw_affiliation_strings":["The University of Texas at Dallas,USA"],"affiliations":[{"raw_affiliation_string":"The University of Texas at Dallas,USA","institution_ids":["https://openalex.org/I162577319"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027962499","display_name":"Tsun-Yu Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tsun-Yu Yang","raw_affiliation_strings":["Duke University,USA"],"affiliations":[{"raw_affiliation_string":"Duke University,USA","institution_ids":["https://openalex.org/I170897317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128072360","display_name":"Zhaoyan Shen","orcid":null},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhaoyan Shen","raw_affiliation_strings":["Shandong University,China"],"affiliations":[{"raw_affiliation_string":"Shandong University,China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102930563","display_name":"Ming-Chang Yang","orcid":"https://orcid.org/0000-0002-4029-757X"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Ming-Chang Yang","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong"],"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5048972267","display_name":"Bingzhe Li","orcid":"https://orcid.org/0000-0002-5815-9706"},"institutions":[{"id":"https://openalex.org/I162577319","display_name":"The University of Texas at Dallas","ror":"https://ror.org/049emcs32","country_code":"US","type":"education","lineage":["https://openalex.org/I162577319"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bingzhe Li","raw_affiliation_strings":["The University of Texas at Dallas,USA"],"affiliations":[{"raw_affiliation_string":"The University of Texas at Dallas,USA","institution_ids":["https://openalex.org/I162577319"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5128126923"],"corresponding_institution_ids":["https://openalex.org/I162577319"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.93154435,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"16"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.26820001006126404,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.26820001006126404,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.15800000727176666,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.1151999980211258,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.631600022315979},{"id":"https://openalex.org/keywords/control-theory","display_name":"Control theory (sociology)","score":0.323199987411499},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.30649998784065247},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.3027999997138977},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.26969999074935913}],"concepts":[{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.631600022315979},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4368000030517578},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36399999260902405},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3402999937534332},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3239000141620636},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.323199987411499},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.30649998784065247},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.3027999997138977},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.26969999074935913},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2606000006198883}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpca68181.2026.11408524","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408524","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Climate action","id":"https://metadata.un.org/sdg/13","score":0.5112784504890442},{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.46492570638656616}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":31,"referenced_works":["https://openalex.org/W114517082","https://openalex.org/W1971630691","https://openalex.org/W2093053744","https://openalex.org/W2101196063","https://openalex.org/W2296407087","https://openalex.org/W2724385931","https://openalex.org/W2964015378","https://openalex.org/W2964571482","https://openalex.org/W3007003693","https://openalex.org/W3015616869","https://openalex.org/W3035233500","https://openalex.org/W3042770487","https://openalex.org/W3157805807","https://openalex.org/W3159109662","https://openalex.org/W3159953606","https://openalex.org/W3206504463","https://openalex.org/W4255949318","https://openalex.org/W4281725510","https://openalex.org/W4281850905","https://openalex.org/W4292718518","https://openalex.org/W4309862248","https://openalex.org/W4319014638","https://openalex.org/W4321479937","https://openalex.org/W4321851522","https://openalex.org/W4372267520","https://openalex.org/W4381832157","https://openalex.org/W4393407046","https://openalex.org/W4393407318","https://openalex.org/W4402042236","https://openalex.org/W7133196215","https://openalex.org/W7133198614"],"related_works":[],"abstract_inverted_index":{"Training":[0],"Graph":[1],"Neural":[2],"Networks":[3],"(GNNs)":[4],"on":[5,41,227],"large-scale":[6,228],"data":[7,64,108,150],"is":[8,156],"essential":[9],"in":[10],"various":[11,208],"applications,":[12],"e.g.,":[13],"transportation,":[14],"and":[15,51,68,81,90,139,179,188,210,223],"molecular":[16],"biology.":[17],"As":[18],"graph":[19,182,229],"sizes":[20],"increasingly":[21],"surpass":[22],"main":[23],"memory":[24,174],"capacities,":[25],"the":[26,60,84,149],"out-of-core":[27],"(OOC)":[28],"GNN":[29,211,225],"training":[30,201,226],"system":[31,105],"(OOC-based)":[32],"has":[33],"been":[34],"proposed,":[35],"a":[36,103,119,180],"scheme":[37],"that":[38,154,185,194],"storing":[39],"graphs":[40],"external":[42],"storage,":[43],"such":[44],"as":[45,70,72],"SSD":[46],"or":[47],"HDD,":[48],"sequentially":[49],"loading":[50,177],"processing":[52],"smaller":[53],"partitions.":[54],"However,":[55],"existing":[56],"OOC-based":[57,88,131],"systems":[58,89],"face":[59],"key":[61,145],"challenges:":[62],"excessive":[63],"migration":[65,109,151],"between":[66],"storage":[67],"memory,":[69],"well":[71],"reduced":[73],"model":[74,114,163],"accuracy.":[75,115,164],"In":[76],"this":[77],"paper,":[78],"we":[79,100,117,168],"theoretically":[80],"empirically":[82],"analyze":[83],"limitations":[85],"of":[86],"state-of-the-art":[87,214],"identify":[91],"opportunities":[92],"for":[93,130,221],"optimization.":[94],"Guided":[95],"by":[96,203],"our":[97],"theoretical":[98],"insights,":[99],"propose":[101,170],"VeloxGNN,":[102],"novel":[104,120],"to":[106,142,205],"improve":[107],"efficiency":[110],"while":[111,161,199],"maintaining":[112,162],"high":[113],"First,":[116],"introduce":[118],"algorithm,":[121],"named":[122],"Delayed":[123],"Gradient":[124],"Propagation":[125],"(DGP),":[126],"which":[127],"specifically":[128],"designed":[129],"system.":[132],"DGP":[133],"leverages":[134],"both":[135],"historical":[136],"node":[137],"embeddings":[138],"unbiased":[140],"gradients":[141],"achieve":[143],"two":[144],"objectives":[146],"simultaneously:":[147],"minimizing":[148],"(by":[152],"ensuring":[153],"dataset":[155],"read":[157],"at":[158],"most":[159],"once)":[160],"To":[165],"support":[166],"DGP,":[167],"then":[169],"system-level":[171],"optimizations:":[172],"dynamic":[173],"management,":[175],"DGP-aware":[176],"order,":[178],"new":[181],"partitioning":[183],"method":[184],"separates":[186],"labeled":[187],"unlabeled":[189],"data.":[190,230],"Experimental":[191],"results":[192],"show":[193],"VeloxGNN":[195],"achieves":[196],"memory-based":[197],"accuracy":[198],"reducing":[200],"time":[202],"17.7%":[204],"73.3%":[206],"across":[207],"datasets":[209],"models,":[212],"outperforming":[213],"methods.":[215],"This":[216],"result":[217],"highlights":[218],"VeloxGNN's":[219],"potential":[220],"efficient":[222],"scalable":[224]},"counts_by_year":[],"updated_date":"2026-03-06T06:45:51.903784","created_date":"2026-03-05T00:00:00"}