{"id":"https://openalex.org/W4395116186","doi":"https://doi.org/10.1145/3620666.3651322","title":"Hector: An Efficient Programming and Compilation Framework for Implementing Relational Graph Neural Networks in GPU Architectures","display_name":"Hector: An Efficient Programming and Compilation Framework for Implementing Relational Graph Neural Networks in GPU Architectures","publication_year":2024,"publication_date":"2024-04-24","ids":{"openalex":"https://openalex.org/W4395116186","doi":"https://doi.org/10.1145/3620666.3651322"},"language":"en","primary_location":{"id":"doi:10.1145/3620666.3651322","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3620666.3651322","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3","raw_type":"proceedings-article"},"type":"conference-paper","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/A5073106039","display_name":"Kun Wu","orcid":"https://orcid.org/0000-0002-0149-1409"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kun Wu","raw_affiliation_strings":["University of Illinois at Urbana-Champaign, Champaign, Illinois, USA"],"raw_orcid":"https://orcid.org/0000-0002-0149-1409","affiliations":[{"raw_affiliation_string":"University of Illinois at Urbana-Champaign, Champaign, Illinois, USA","institution_ids":["https://openalex.org/I157725225"]}]},{"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/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mert Hidayeto\u011flu","raw_affiliation_strings":["Stanford University, Stanford, California, USA"],"raw_orcid":"https://orcid.org/0000-0001-9276-5075","affiliations":[{"raw_affiliation_string":"Stanford University, Stanford, California, USA","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101675531","display_name":"Xiang Song","orcid":"https://orcid.org/0000-0001-5030-5054"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiang Song","raw_affiliation_strings":["AWS AI, Santa Clara, California, USA","of at - USA","of , USA"],"raw_orcid":"https://orcid.org/0000-0001-5030-5054","affiliations":[{"raw_affiliation_string":"AWS AI, Santa Clara, California, USA","institution_ids":[]},{"raw_affiliation_string":"of at - USA","institution_ids":[]},{"raw_affiliation_string":"of , USA","institution_ids":[]}]},{"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/I204250578","display_name":"University of California, Irvine","ror":"https://ror.org/04gyf1771","country_code":"US","type":"education","lineage":["https://openalex.org/I204250578"]},{"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":"Sitao Huang","raw_affiliation_strings":["University of California, Irvine, Irvine, California, United States of America","of , USA","of at - USA","Nvidia of - USA"],"raw_orcid":"https://orcid.org/0000-0001-7669-1467","affiliations":[{"raw_affiliation_string":"University of California, Irvine, Irvine, California, United States of America","institution_ids":["https://openalex.org/I204250578"]},{"raw_affiliation_string":"of , USA","institution_ids":[]},{"raw_affiliation_string":"of at - USA","institution_ids":[]},{"raw_affiliation_string":"Nvidia of - USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060763203","display_name":"Da Zheng","orcid":"https://orcid.org/0000-0001-8115-5415"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Da Zheng","raw_affiliation_strings":["AWS AI, Santa Clara, California, USA","of , USA","of at - USA"],"raw_orcid":"https://orcid.org/0000-0001-8115-5415","affiliations":[{"raw_affiliation_string":"AWS AI, Santa Clara, California, USA","institution_ids":[]},{"raw_affiliation_string":"of , USA","institution_ids":[]},{"raw_affiliation_string":"of at - USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063948428","display_name":"Israt Nisa","orcid":"https://orcid.org/0000-0001-5022-5716"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Israt Nisa","raw_affiliation_strings":["AWS AI, Santa Clara, California, USA"],"raw_orcid":"https://orcid.org/0000-0001-5022-5716","affiliations":[{"raw_affiliation_string":"AWS AI, Santa Clara, California, USA","institution_ids":[]}]},{"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/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]},{"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, Champaign, Illinois, United States of America","University of Illinois at Urbana-Champaign, Champaign, Illinois, USA"],"raw_orcid":"https://orcid.org/0000-0003-2532-5349","affiliations":[{"raw_affiliation_string":"Nvidia, Champaign, Illinois, United States of America","institution_ids":["https://openalex.org/I4210127875"]},{"raw_affiliation_string":"University of Illinois at Urbana-Champaign, Champaign, Illinois, USA","institution_ids":["https://openalex.org/I157725225"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"528","last_page":"544"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9998999834060669,"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":0.9998999834060669,"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/T12292","display_name":"Graph Theory and Algorithms","score":0.9904000163078308,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9894000291824341,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8575886487960815},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.5040799379348755},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.4926515221595764},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.48100218176841736},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.4616803526878357},{"id":"https://openalex.org/keywords/programming-paradigm","display_name":"Programming paradigm","score":0.45071256160736084},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4469718039035797},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4426478147506714},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3833639919757843},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.26383161544799805}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8575886487960815},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.5040799379348755},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.4926515221595764},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.48100218176841736},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.4616803526878357},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.45071256160736084},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4469718039035797},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4426478147506714},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3833639919757843},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.26383161544799805},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"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/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3620666.3651322","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3620666.3651322","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W1506109531","https://openalex.org/W1849400742","https://openalex.org/W2056562706","https://openalex.org/W2140119009","https://openalex.org/W2172654076","https://openalex.org/W2245493112","https://openalex.org/W2250184916","https://openalex.org/W2590246587","https://openalex.org/W2604314403","https://openalex.org/W2804032941","https://openalex.org/W3012871709","https://openalex.org/W3080508754","https://openalex.org/W3115410382","https://openalex.org/W3122286897","https://openalex.org/W3138871094","https://openalex.org/W3158027451","https://openalex.org/W3159953606","https://openalex.org/W3198239267","https://openalex.org/W3213084269","https://openalex.org/W4255450819","https://openalex.org/W4283377648","https://openalex.org/W4327911434","https://openalex.org/W4385585441"],"related_works":["https://openalex.org/W2115140794","https://openalex.org/W2041120224","https://openalex.org/W1504320321","https://openalex.org/W2067902980","https://openalex.org/W2548529098","https://openalex.org/W4312275919","https://openalex.org/W1493568480","https://openalex.org/W2338293335","https://openalex.org/W2039378765","https://openalex.org/W2614685449"],"abstract_inverted_index":{"Relational":[0],"graph":[1,6],"neural":[2,7],"networks":[3,8],"(RGNNs)":[4],"are":[5],"with":[9,72,118,173],"dedicated":[10],"structures":[11],"for":[12],"modeling":[13],"the":[14,51,54,98,174,225,235,248],"different":[15],"types":[16],"of":[17,101,234,237,245,255,258],"nodes":[18],"and":[19,38,43,57,60,75,90,104,113,128,135,154,167,183,195,219,251,260,277,296],"edges":[20],"in":[21,30,110,165,170,210,242,282],"heterogeneous":[22,187],"graphs.":[23],"While":[24],"RGNNs":[25],"have":[26],"been":[27],"increasingly":[28],"adopted":[29],"many":[31],"real-world":[32],"applications":[33],"due":[34],"to":[35,65,106,123,141,162,222,229,273],"their":[36,70],"versatility":[37],"accuracy,":[39],"they":[40],"pose":[41],"performance":[42],"system":[44,226],"design":[45],"challenges:":[46],"inherent":[47],"memory-intensive":[48],"computation":[49],"patterns,":[50],"gap":[52],"between":[53],"programming":[55,62,143,238],"interface":[56],"kernel":[58],"APIs,":[59],"heavy":[61],"effort":[63],"required":[64],"optimize":[66],"kernels":[67],"caused":[68],"by":[69,190,227,290,293],"coupling":[71],"data":[73,120,126,133],"layout":[74],"heterogeneity.":[76],"To":[77],"systematically":[78],"address":[79],"these":[80,211],"challenges,":[81],"we":[82,265],"propose":[83,215],"Hector,":[84],"a":[85,155,253],"novel":[86],"two-level":[87],"intermediate":[88],"representation":[89],"its":[91],"code":[92,117,246],"generator":[93],"framework":[94],"that":[95,267],"(a)":[96],"captures":[97],"key":[99],"properties":[100],"RGNN":[102],"models,":[103,180],"opportunities":[105],"reduce":[107,142],"memory":[108,269],"accesses":[109],"inter-operator":[111],"scheduling":[112],"materialization,":[114],"(b)":[115],"generates":[116,252],"flexible":[119],"access":[121],"schemes":[122],"eliminate":[124],"redundant":[125],"copies,":[127],"(c)":[129],"decouples":[130],"model":[131],"semantics,":[132],"layout,":[134],"operators-specific":[136],"optimizations":[137],"from":[138],"each":[139],"other":[140],"effort.":[144],"By":[145],"building":[146],"on":[147,178],"one":[148],"general":[149],"matrix":[150],"multiply":[151],"(GEMM)":[152],"template":[153],"node/edge":[156],"traversal":[157],"template,":[158],"Hector":[159,202,240,272],"achieves":[160],"up":[161,228],"9.9\u00d7":[163],"speed-up":[164,169],"inference":[166],"43.7\u00d7":[168],"training":[171],"compared":[172],"state-of-the-art":[175],"public":[176],"systems":[177],"select":[179],"RGCN,":[181],"RGAT":[182],"HGT,":[184],"when":[185],"running":[186],"graphs":[188],"provided":[189],"Deep":[191],"Graph":[192,197],"Library":[193],"(DGL)":[194],"Open":[196],"Benchmark":[198],"(OGB).":[199],"In":[200],"addition,":[201],"does":[203],"not":[204],"trigger":[205],"any":[206],"out-of-memory":[207],"(OOM)":[208],"exception":[209],"tests.":[212],"We":[213],"also":[214],"linear":[216],"operator":[217],"reordering":[218],"compact":[220],"materialization":[221],"further":[223],"accelerate":[224],"3.8\u00d7.":[230],"As":[231],"an":[232],"indicator":[233],"reduction":[236],"effort,":[239],"takes":[241],"51":[243],"lines":[244,257],"expressing":[247],"three":[249],"models":[250],"total":[254],"8K":[256],"CUDA":[259],"C++":[261],"code.":[262],"Through":[263],"profiling,":[264],"found":[266],"higher":[268,280],"efficiency":[270],"allows":[271],"accommodate":[274],"larger":[275],"input":[276],"therefore":[278],"attain":[279],"throughput":[281],"forward":[283],"propagation,":[284],"while":[285],"backward":[286],"propagation":[287],"is":[288],"bound":[289],"latency":[291],"introduced":[292],"atomic":[294],"updates":[295],"outer":[297],"products.":[298]},"counts_by_year":[{"year":2025,"cited_by_count":3}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}