{"id":"https://openalex.org/W4391682413","doi":"https://doi.org/10.1145/3644712","title":"TLPGNN: A Lightweight Two-level Parallelism Paradigm for Graph Neural Network Computation on Single and Multiple GPUs","display_name":"TLPGNN: A Lightweight Two-level Parallelism Paradigm for Graph Neural Network Computation on Single and Multiple GPUs","publication_year":2024,"publication_date":"2024-02-09","ids":{"openalex":"https://openalex.org/W4391682413","doi":"https://doi.org/10.1145/3644712"},"language":"en","primary_location":{"id":"doi:10.1145/3644712","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3644712","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3644712","source":{"id":"https://openalex.org/S2483380313","display_name":"ACM Transactions on Parallel Computing","issn_l":"2329-4949","issn":["2329-4949","2329-4957"],"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":"ACM Transactions on Parallel Computing","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3644712","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101992907","display_name":"Qiang Fu","orcid":"https://orcid.org/0009-0004-7159-6515"},"institutions":[{"id":"https://openalex.org/I4210137977","display_name":"Advanced Micro Devices (United States)","ror":"https://ror.org/04kd6c783","country_code":"US","type":"company","lineage":["https://openalex.org/I4210137977"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qiang Fu","raw_affiliation_strings":["Advanced Micro Devices Inc Austin, Austin, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Advanced Micro Devices Inc Austin, Austin, USA","institution_ids":["https://openalex.org/I4210137977"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066436066","display_name":"Yuede Ji","orcid":"https://orcid.org/0000-0002-2419-6592"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yuede Ji","raw_affiliation_strings":["Computer Science and Engineering, University of North Texas, Denton, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Computer Science and Engineering, University of North Texas, Denton, USA","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026169135","display_name":"Thomas B. Rolinger","orcid":"https://orcid.org/0000-0001-8383-4737"},"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":"Thomas Rolinger","raw_affiliation_strings":["NVIDIA, Austin, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"NVIDIA, Austin, USA","institution_ids":["https://openalex.org/I4210127875"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002254350","display_name":"H. Howie Huang","orcid":"https://orcid.org/0000-0001-8588-7680"},"institutions":[{"id":"https://openalex.org/I193531525","display_name":"George Washington University","ror":"https://ror.org/00y4zzh67","country_code":"US","type":"education","lineage":["https://openalex.org/I193531525"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"H. Howie Huang","raw_affiliation_strings":["The George Washington University, Washington, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The George Washington University, Washington, USA","institution_ids":["https://openalex.org/I193531525"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.4087,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.55331223,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":"11","issue":"2","first_page":"1","last_page":"28"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12292","display_name":"Graph Theory and Algorithms","score":0.9997000098228455,"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":0.9997000098228455,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.9994999766349792,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.9836000204086304,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials 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":0.8614989519119263},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.6653154492378235},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6503301858901978},{"id":"https://openalex.org/keywords/data-parallelism","display_name":"Data parallelism","score":0.6383197903633118},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.5570874810218811},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.4923797845840454},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.46801531314849854},{"id":"https://openalex.org/keywords/instruction-level-parallelism","display_name":"Instruction-level parallelism","score":0.4532897472381592},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.44629156589508057},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4371013641357422},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.42142415046691895},{"id":"https://openalex.org/keywords/parallelism","display_name":"Parallelism (grammar)","score":0.3968597948551178},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3762739300727844},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.35124653577804565},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.33094722032546997},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.14656254649162292}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8614989519119263},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6653154492378235},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6503301858901978},{"id":"https://openalex.org/C61483411","wikidata":"https://www.wikidata.org/wiki/Q3124522","display_name":"Data parallelism","level":3,"score":0.6383197903633118},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.5570874810218811},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.4923797845840454},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.46801531314849854},{"id":"https://openalex.org/C140763907","wikidata":"https://www.wikidata.org/wiki/Q2714055","display_name":"Instruction-level parallelism","level":3,"score":0.4532897472381592},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.44629156589508057},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4371013641357422},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.42142415046691895},{"id":"https://openalex.org/C2781172179","wikidata":"https://www.wikidata.org/wiki/Q853109","display_name":"Parallelism (grammar)","level":2,"score":0.3968597948551178},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3762739300727844},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.35124653577804565},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.33094722032546997},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.14656254649162292},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","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},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3644712","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3644712","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3644712","source":{"id":"https://openalex.org/S2483380313","display_name":"ACM Transactions on Parallel Computing","issn_l":"2329-4949","issn":["2329-4949","2329-4957"],"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":"ACM Transactions on Parallel Computing","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1145/3644712","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3644712","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3644712","source":{"id":"https://openalex.org/S2483380313","display_name":"ACM Transactions on Parallel Computing","issn_l":"2329-4949","issn":["2329-4949","2329-4957"],"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":"ACM Transactions on Parallel Computing","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4391682413.pdf","grobid_xml":"https://content.openalex.org/works/W4391682413.grobid-xml"},"referenced_works_count":36,"referenced_works":["https://openalex.org/W1955594754","https://openalex.org/W1965830721","https://openalex.org/W2000041758","https://openalex.org/W2001575213","https://openalex.org/W2034102265","https://openalex.org/W2053076698","https://openalex.org/W2080098453","https://openalex.org/W2162630660","https://openalex.org/W2170616854","https://openalex.org/W2250844151","https://openalex.org/W2484446135","https://openalex.org/W2718955078","https://openalex.org/W2807021761","https://openalex.org/W2902940306","https://openalex.org/W2914097545","https://openalex.org/W2914721378","https://openalex.org/W2943605315","https://openalex.org/W2945827377","https://openalex.org/W2951135776","https://openalex.org/W2962865652","https://openalex.org/W2963757395","https://openalex.org/W3004507689","https://openalex.org/W3037631590","https://openalex.org/W3037699692","https://openalex.org/W3042370959","https://openalex.org/W3100848837","https://openalex.org/W3101553402","https://openalex.org/W3103513278","https://openalex.org/W3125542198","https://openalex.org/W3132185085","https://openalex.org/W3138165103","https://openalex.org/W3158027451","https://openalex.org/W3173204484","https://openalex.org/W3201678711","https://openalex.org/W4285378361","https://openalex.org/W6748856961"],"related_works":["https://openalex.org/W2950520577","https://openalex.org/W2003935582","https://openalex.org/W1991844655","https://openalex.org/W2105992728","https://openalex.org/W1229628","https://openalex.org/W74409296","https://openalex.org/W2009213655","https://openalex.org/W2593878938","https://openalex.org/W2494130044","https://openalex.org/W3104857564"],"abstract_inverted_index":{"Graph":[0,237],"Neural":[1],"Networks":[2],"(GNNs)":[3],"are":[4],"an":[5,198],"emerging":[6],"class":[7],"of":[8,25,35,54,76,108,113,120,174,223,245,254],"deep":[9],"learning":[10],"models":[11],"specifically":[12],"designed":[13],"for":[14,70,99,140,147],"graph-structured":[15],"data.":[16],"They":[17],"have":[18,63],"been":[19,64],"effectively":[20],"employed":[21],"in":[22],"a":[23,94,105,153,206,267],"variety":[24],"real-world":[26],"applications,":[27],"including":[28,235],"recommendation":[29],"systems,":[30,234],"drug":[31],"development,":[32],"and":[33,46,83,144,177,241,248,266],"analysis":[34,107],"social":[36],"networks.":[37],"The":[38],"GNN":[39,100,114,121,132,232],"computation":[40,57,69,133,233],"includes":[41],"regular":[42],"neural":[43],"network":[44],"operations":[45],"general":[47],"graph":[48],"convolution":[49],"operations,":[50,82],"which":[51],"take":[52],"most":[53],"the":[55,68,74,109,118,125,131,141,148,161,168,172,179,221],"total":[56],"time.":[58],"Though":[59],"several":[60],"recent":[61],"works":[62],"proposed":[65],"to":[66,116,159,170,185,193,204],"accelerate":[67],"GNNs,":[71],"they":[72],"face":[73],"limitations":[75],"heavy":[77],"pre-processing,":[78],"low":[79],"efficiency":[80],"atomic":[81],"unnecessary":[84,187],"kernel":[85,175],"launches.":[86],"In":[87],"this":[88],"article,":[89],"we":[90,103,128,151,166,196],"design":[91],"TLPGNN":[92,192,256],",":[93],"lightweight":[95],"two-level":[96],"parallelism":[97,139,146],"paradigm":[98],"computation.":[101],"First,":[102],"conduct":[104],"systematic":[106],"hardware":[110],"resource":[111],"usage":[112],"workloads":[115,122],"understand":[117],"characteristics":[119],"deeply.":[123],"With":[124],"insightful":[126],"observations,":[127],"then":[129],"divide":[130],"into":[134,183],"two":[135],"levels,":[136],"i.e.,":[137],"vertex":[138],"first":[142],"level":[143],"feature":[145],"second.":[149],"Next,":[150],"employ":[152],"novel":[154],"hybrid":[155],"dynamic":[156],"workload":[157,163,208,269],"assignment":[158],"address":[160],"imbalanced":[162],"distribution.":[164,270],"Furthermore,":[165],"fuse":[167],"kernels":[169],"reduce":[171],"number":[173],"launches":[176],"cache":[178],"frequently":[180],"accessed":[181],"data":[182],"registers":[184],"avoid":[186],"memory":[188],"traffic.":[189],"To":[190],"scale":[191],"multi-GPU":[194],"environments,":[195],"propose":[197],"edge-aware":[199],"row-wise":[200],"1-D":[201],"partition":[202],"method":[203],"ensure":[205],"balanced":[207],"distribution":[209],"across":[210],"different":[211],"GPU":[212],"devices.":[213],"Experimental":[214],"results":[215],"on":[216,251],"various":[217],"benchmark":[218],"datasets":[219],"demonstrate":[220,258],"superiority":[222],"our":[224,260],"approach,":[225],"achieving":[226],"substantial":[227],"performance":[228],"improvement":[229],"over":[230],"state-of-the-art":[231],"Deep":[236],"Library":[238],"(DGL),":[239],"GNNAdvisor,":[240],"FeatGraph,":[242],"with":[243],"speedups":[244],"6.1\u00d7,":[246],"7.7\u00d7,":[247],"3.0\u00d7,":[249],"respectively,":[250],"average.":[252],"Evaluations":[253],"multiple-GPU":[255],"also":[257],"that":[259],"solution":[261],"achieves":[262],"both":[263],"linear":[264],"scalability":[265],"well-balanced":[268]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}