{"id":"https://openalex.org/W4407214500","doi":"https://doi.org/10.1109/lca.2025.3539371","title":"Comprehensive Design Space Exploration for Graph Neural Network Aggregation on GPUs","display_name":"Comprehensive Design Space Exploration for Graph Neural Network Aggregation on GPUs","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4407214500","doi":"https://doi.org/10.1109/lca.2025.3539371"},"language":"en","primary_location":{"id":"doi:10.1109/lca.2025.3539371","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lca.2025.3539371","pdf_url":null,"source":{"id":"https://openalex.org/S17643076","display_name":"IEEE Computer Architecture Letters","issn_l":"1556-6056","issn":["1556-6056","1556-6064","2473-2575"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Computer Architecture Letters","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":null,"display_name":"Hyunwoo Nam","orcid":"https://orcid.org/0009-0002-7774-8028"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Hyunwoo Nam","raw_affiliation_strings":["Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087954884","display_name":"Jay Hwan Lee","orcid":"https://orcid.org/0000-0002-1975-2363"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jay Hwan Lee","raw_affiliation_strings":["Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012493068","display_name":"Shinhyung Yang","orcid":"https://orcid.org/0000-0002-8997-9942"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Shinhyung Yang","raw_affiliation_strings":["Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea","Department of Computer Science and Engineering, Yonsei University, South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea","institution_ids":["https://openalex.org/I193775966"]},{"raw_affiliation_string":"Department of Computer Science and Engineering, Yonsei University, South Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101489113","display_name":"Yeonsoo Kim","orcid":"https://orcid.org/0000-0002-9437-4665"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yeonsoo Kim","raw_affiliation_strings":["Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006654611","display_name":"Jiun Jeong","orcid":null},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jiun Jeong","raw_affiliation_strings":["Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101846451","display_name":"Jeong-Geun Kim","orcid":"https://orcid.org/0000-0003-0264-0979"},"institutions":[{"id":"https://openalex.org/I31419693","display_name":"Kyungpook National University","ror":"https://ror.org/040c17130","country_code":"KR","type":"education","lineage":["https://openalex.org/I31419693"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jeonggeun Kim","raw_affiliation_strings":["School of Computer Science and Engineering, Kyungpook National University, Daegu, South Korea"],"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Kyungpook National University, Daegu, South Korea","institution_ids":["https://openalex.org/I31419693"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5012072107","display_name":"Bernd Burgstaller","orcid":"https://orcid.org/0000-0002-0374-8853"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Bernd Burgstaller","raw_affiliation_strings":["Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Yonsei University, Seoul, South Korea","institution_ids":["https://openalex.org/I193775966"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I193775966"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00820825,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"24","issue":"1","first_page":"45","last_page":"48"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9807000160217285,"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":0.9807000160217285,"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":0.9661999940872192,"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/T10320","display_name":"Neural Networks and Applications","score":0.9517999887466431,"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/computer-science","display_name":"Computer science","score":0.8121523857116699},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.5516442656517029},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5295649170875549},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5040261149406433},{"id":"https://openalex.org/keywords/space-exploration","display_name":"Space exploration","score":0.4642171859741211},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4476708769798279},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.41926199197769165},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.36373651027679443},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.34205925464630127},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.27586832642555237},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.14721491932868958},{"id":"https://openalex.org/keywords/aerospace-engineering","display_name":"Aerospace engineering","score":0.10963377356529236}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8121523857116699},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.5516442656517029},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5295649170875549},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5040261149406433},{"id":"https://openalex.org/C104060986","wikidata":"https://www.wikidata.org/wiki/Q180046","display_name":"Space exploration","level":2,"score":0.4642171859741211},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4476708769798279},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.41926199197769165},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.36373651027679443},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.34205925464630127},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.27586832642555237},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.14721491932868958},{"id":"https://openalex.org/C146978453","wikidata":"https://www.wikidata.org/wiki/Q3798668","display_name":"Aerospace engineering","level":1,"score":0.10963377356529236},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/lca.2025.3539371","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lca.2025.3539371","pdf_url":null,"source":{"id":"https://openalex.org/S17643076","display_name":"IEEE Computer Architecture Letters","issn_l":"1556-6056","issn":["1556-6056","1556-6064","2473-2575"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Computer Architecture Letters","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G7296649959","display_name":null,"funder_award_id":"IO201209-07887-01","funder_id":"https://openalex.org/F4320332195","funder_display_name":"Samsung"}],"funders":[{"id":"https://openalex.org/F4320332195","display_name":"Samsung","ror":"https://ror.org/04w3jy968"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":17,"referenced_works":["https://openalex.org/W1512387364","https://openalex.org/W2979340153","https://openalex.org/W2990045899","https://openalex.org/W3004208721","https://openalex.org/W3011667710","https://openalex.org/W3082448434","https://openalex.org/W3102510044","https://openalex.org/W3105753905","https://openalex.org/W3130554079","https://openalex.org/W4283377648","https://openalex.org/W4312657855","https://openalex.org/W4387789707","https://openalex.org/W6699364125","https://openalex.org/W6726873649","https://openalex.org/W6738964360","https://openalex.org/W6760045743","https://openalex.org/W6765543928"],"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/W2338293335","https://openalex.org/W2039378765","https://openalex.org/W2614685449","https://openalex.org/W283879968"],"abstract_inverted_index":{"Graph":[0],"neural":[1],"networks":[2],"(GNNs)":[3],"have":[4,70],"become":[5,27],"the":[6,24,28,53,81,102,110,118,131],"state-of-the-art":[7],"technology":[8],"for":[9,31,101,114,161],"extracting":[10],"and":[11,34,43,142],"predicting":[12],"data":[13],"representations":[14],"on":[15,61,169],"graphs.":[16],"With":[17],"increasing":[18],"demand":[19],"to":[20,73,97,164],"accelerate":[21],"GNN":[22,32,68,167],"computations,":[23],"GPU":[25,99,111],"has":[26],"dominant":[29],"platform":[30],"training":[33],"inference.":[35],"GNNs":[36],"consist":[37],"of":[38,52,83,90,122,125,134],"a":[39,44,57,87,123,143],"compute-bound":[40],"combination":[41],"phase":[42,55],"memory-bound":[45],"aggregation":[46,54,103,115,135,168],"phase.":[47,104],"The":[48],"memory":[49,147],"access":[50,148],"patterns":[51],"remain":[56,151],"major":[58],"performance":[59,119,137],"bottleneck":[60],"GPUs,":[62],"despite":[63],"recent":[64],"microarchitectural":[65],"enhancements.":[66],"Although":[67],"characterizations":[69],"been":[71],"conducted":[72],"investigate":[74],"this":[75,106],"bottleneck,":[76],"they":[77],"did":[78],"not":[79],"reveal":[80],"impact":[82],"architectural":[84,126],"modifications.":[85,127],"However,":[86],"comprehensive":[88],"understanding":[89],"improvements":[91],"from":[92],"such":[93],"modifications":[94],"is":[95],"imperative":[96],"devise":[98],"optimizations":[100,163],"In":[105],"letter,":[107],"we":[108],"explore":[109],"design":[112],"space":[113],"by":[116],"assessing":[117],"improvement":[120],"potential":[121],"series":[124],"We":[128],"find":[129],"that":[130],"low":[132],"locality":[133],"deteriorates":[136],"with":[138,154],"increased":[139],"thread-level":[140],"parallelism,":[141],"significant":[144],"enhancement":[145],"follows":[146],"optimizations,":[149],"which":[150],"effective":[152],"even":[153],"software":[155],"optimization.":[156],"Our":[157],"analysis":[158],"provides":[159],"insights":[160],"hardware":[162],"significantly":[165],"improve":[166],"GPUs.":[170]},"counts_by_year":[],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}