{"id":"https://openalex.org/W4396601595","doi":"https://doi.org/10.14778/3648160.3648166","title":"Accelerating Sampling and Aggregation Operations in GNN Frameworks with GPU Initiated Direct Storage Accesses","display_name":"Accelerating Sampling and Aggregation Operations in GNN Frameworks with GPU Initiated Direct Storage Accesses","publication_year":2024,"publication_date":"2024-02-01","ids":{"openalex":"https://openalex.org/W4396601595","doi":"https://doi.org/10.14778/3648160.3648166"},"language":"en","primary_location":{"id":"doi:10.14778/3648160.3648166","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3648160.3648166","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":"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/A5100776534","display_name":"Jeongmin Park","orcid":"https://orcid.org/0000-0001-8027-0876"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jeongmin Brian Park","raw_affiliation_strings":["UIUC, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UIUC, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041724753","display_name":"Vikram Sharma Mailthody","orcid":"https://orcid.org/0000-0002-9611-8075"},"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":"Vikram Sharma Mailthody","raw_affiliation_strings":["NVIDIA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046830079","display_name":"Zaid Qureshi","orcid":"https://orcid.org/0000-0003-1766-1289"},"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":"Zaid Qureshi","raw_affiliation_strings":["NVIDIA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"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/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":"Wen-mei Hwu","raw_affiliation_strings":["NVIDIA/UIUC, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA/UIUC, USA","institution_ids":["https://openalex.org/I4210127875"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":4.3935,"has_fulltext":false,"cited_by_count":22,"citation_normalized_percentile":{"value":0.94794897,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":100},"biblio":{"volume":"17","issue":"6","first_page":"1227","last_page":"1240"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9962999820709229,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9962999820709229,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9936000108718872,"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":0.9933000206947327,"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/computer-science","display_name":"Computer science","score":0.7219685316085815},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5135138630867004},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.4965510964393616},{"id":"https://openalex.org/keywords/computer-data-storage","display_name":"Computer data storage","score":0.4103803336620331},{"id":"https://openalex.org/keywords/database","display_name":"Database","score":0.32937949895858765},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.2887028455734253}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7219685316085815},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5135138630867004},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.4965510964393616},{"id":"https://openalex.org/C194739806","wikidata":"https://www.wikidata.org/wiki/Q66221","display_name":"Computer data storage","level":2,"score":0.4103803336620331},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.32937949895858765},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2887028455734253},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.0},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.14778/3648160.3648166","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3648160.3648166","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"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":36,"referenced_works":["https://openalex.org/W1493774699","https://openalex.org/W2053425642","https://openalex.org/W2154987621","https://openalex.org/W2914304175","https://openalex.org/W2914721378","https://openalex.org/W2945827377","https://openalex.org/W2945996535","https://openalex.org/W2970090796","https://openalex.org/W2970929262","https://openalex.org/W3002924435","https://openalex.org/W3022945404","https://openalex.org/W3036016037","https://openalex.org/W3037699692","https://openalex.org/W3040478789","https://openalex.org/W3086866381","https://openalex.org/W3096566397","https://openalex.org/W3099825604","https://openalex.org/W3101553402","https://openalex.org/W3104001151","https://openalex.org/W3104307750","https://openalex.org/W3147562045","https://openalex.org/W3159506417","https://openalex.org/W3159953606","https://openalex.org/W3167128805","https://openalex.org/W3175275734","https://openalex.org/W3176287504","https://openalex.org/W3205803342","https://openalex.org/W3207233976","https://openalex.org/W4281725510","https://openalex.org/W4281935531","https://openalex.org/W4282576620","https://openalex.org/W4282593722","https://openalex.org/W4290927864","https://openalex.org/W4292718518","https://openalex.org/W4318541614","https://openalex.org/W4321446145"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W2382290278","https://openalex.org/W4395014643","https://openalex.org/W4391913857","https://openalex.org/W2350741829"],"abstract_inverted_index":{"Graph":[0],"Neural":[1],"Networks":[2],"(GNNs)":[3],"are":[4,79],"emerging":[5],"as":[6,168],"a":[7,41,197,234],"powerful":[8],"tool":[9],"for":[10,55,60,65,126,158,226],"learning":[11],"from":[12,192],"graph-structured":[13],"data":[14],"and":[15,52,68,74,94,108,134,172,211],"performing":[16],"sophisticated":[17],"inference":[18],"tasks":[19],"in":[20,114],"various":[21],"application":[22],"domains.":[23],"Although":[24],"GNNs":[25,62],"have":[26],"been":[27],"shown":[28],"to":[29,45,99,153,187,218,255,259],"be":[30],"effective":[31],"on":[32,37,81,237],"modest-sized":[33],"graphs,":[34],"training":[35,61,73,157,229,251],"them":[36],"large-scale":[38,159],"graphs":[39,160],"remains":[40],"significant":[42],"challenge":[43],"due":[44],"the":[46,72,91,106,115,118,122,145,203,223,243,247,260],"lack":[47],"of":[48,76,132,199],"efficient":[49],"storage":[50,205],"access":[51,206],"caching":[53],"methods":[54],"graph":[56,66,92,107],"data.":[57],"Existing":[58],"frameworks":[59],"use":[63],"CPUs":[64,88],"sampling":[67,93],"feature":[69,95,190],"aggregation,":[70],"while":[71,161],"updating":[75],"model":[77],"weights":[78],"executed":[80],"GPUs.":[82,103],"However,":[83],"our":[84],"in-depth":[85],"profiling":[86],"shows":[87,241],"cannot":[89],"achieve":[90],"aggregation":[96],"throughput":[97],"required":[98],"keep":[100],"up":[101,254],"with":[102,215],"Furthermore,":[104],"when":[105,257],"its":[109],"embeddings":[110],"do":[111],"not":[112],"fit":[113],"CPU":[116,169,209],"memory,":[117,170],"overhead":[119],"introduced":[120],"by":[121,183,253],"operating":[123],"system,":[124],"say":[125],"handling":[127],"page-faults,":[128],"causes":[129],"gross":[130],"under-utilization":[131],"hardware":[133,165],"prolonged":[135],"end-to-end":[136,228],"execution":[137],"time.":[138],"To":[139],"address":[140],"these":[141],"issues,":[142],"we":[143,195],"propose":[144],"GPU":[146,173,185,212,236],"Initiated":[147],"Direct":[148],"Storage":[149],"Access":[150],"(GIDS)":[151],"dataloader,":[152],"enable":[154],"GPU-oriented":[155],"GNN":[156,239,250],"efficiently":[162],"utilizing":[163],"all":[164],"resources,":[166],"such":[167],"storage,":[171],"memory.":[174],"The":[175],"GIDS":[176,244],"dataloader":[177,245],"first":[178],"addresses":[179],"memory":[180],"capacity":[181],"constraints":[182],"enabling":[184],"threads":[186],"directly":[188],"fetch":[189],"vectors":[191],"storage.":[193],"Then,":[194],"introduce":[196],"set":[198],"innovative":[200],"solutions,":[201],"including":[202],"dynamic":[204],"accumulator,":[207],"constant":[208],"buffer,":[210],"software":[213],"cache":[214],"window":[216],"buffering,":[217],"balance":[219],"resource":[220],"utilization":[221],"across":[222],"entire":[224],"system":[225],"improved":[227],"throughput.":[230],"Our":[231],"evaluation":[232],"using":[233],"single":[235],"terabyte-scale":[238],"datasets":[240],"that":[242],"accelerates":[246],"overall":[248],"DGL":[249,263],"pipeline":[252],"582\u00d7":[256],"compared":[258],"current,":[261],"state-of-the-art":[262],"dataloader.":[264]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":17},{"year":2024,"cited_by_count":1}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}