{"id":"https://openalex.org/W4205993941","doi":"https://doi.org/10.1109/tpds.2021.3133691","title":"SGCNAX: A Scalable Graph Convolutional Neural Network Accelerator with Workload Balancing","display_name":"SGCNAX: A Scalable Graph Convolutional Neural Network Accelerator with Workload Balancing","publication_year":2021,"publication_date":"2021-01-01","ids":{"openalex":"https://openalex.org/W4205993941","doi":"https://doi.org/10.1109/tpds.2021.3133691"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2021.3133691","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2021.3133691","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Transactions on Parallel and Distributed Systems","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/A5100352041","display_name":"Jiajun Li","orcid":"https://orcid.org/0000-0002-7208-9345"},"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":"Jiajun Li","raw_affiliation_strings":["Electrical and Computer Engineering, The George Washington University, 8367 ARLINGTON, District of Columbia, United States, 22202 (e-mail: lijiajun@gwu.edu)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Electrical and Computer Engineering, The George Washington University, 8367 ARLINGTON, District of Columbia, United States, 22202 (e-mail: lijiajun@gwu.edu)","institution_ids":["https://openalex.org/I193531525"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029155710","display_name":"Hao Zheng","orcid":"https://orcid.org/0000-0003-4391-2774"},"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":"Hao Zheng","raw_affiliation_strings":["Department of Electrical and Computer Engineering, George Washington University, 8367 Washington, District of Columbia, United States, 20052-0086 (e-mail: haozheng@gwu.edu)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, George Washington University, 8367 Washington, District of Columbia, United States, 20052-0086 (e-mail: haozheng@gwu.edu)","institution_ids":["https://openalex.org/I193531525"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100360174","display_name":"Ke Wang","orcid":"https://orcid.org/0009-0008-4869-5135"},"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":"Ke Wang","raw_affiliation_strings":["School of Engineering and Applied Science, George Washington University, Washington, District of Columbia, United States, 20052 (e-mail: cory@gwu.edu)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Engineering and Applied Science, George Washington University, Washington, District of Columbia, United States, 20052 (e-mail: cory@gwu.edu)","institution_ids":["https://openalex.org/I193531525"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5034189643","display_name":"Ahmed Louri","orcid":"https://orcid.org/0000-0003-4262-6688"},"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":"Ahmed Louri","raw_affiliation_strings":["ECE, The George Washington University, 8367 Washington, District of Columbia, United States, (e-mail: louri@gwu.edu)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ECE, The George Washington University, 8367 Washington, District of Columbia, United States, (e-mail: louri@gwu.edu)","institution_ids":["https://openalex.org/I193531525"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I193531525"],"apc_list":null,"apc_paid":null,"fwci":2.3202,"has_fulltext":false,"cited_by_count":27,"citation_normalized_percentile":{"value":0.90361765,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"1"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"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"}},"topics":[{"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9968000054359436,"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"}},{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9810000061988831,"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":0.8396624326705933},{"id":"https://openalex.org/keywords/dataflow","display_name":"Dataflow","score":0.7601034641265869},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.7137151956558228},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6342555284500122},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.6198650598526001},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.5252877473831177},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4527982175350189},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.4370258152484894},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.4257334768772125},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.41405433416366577},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.36007869243621826},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3313751816749573},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.2568341791629791},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.23282349109649658},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.16580694913864136},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.11652687191963196},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.1043052077293396}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8396624326705933},{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.7601034641265869},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7137151956558228},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6342555284500122},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.6198650598526001},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.5252877473831177},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4527982175350189},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4370258152484894},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.4257334768772125},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.41405433416366577},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.36007869243621826},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3313751816749573},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2568341791629791},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.23282349109649658},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.16580694913864136},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.11652687191963196},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.1043052077293396},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2021.3133691","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2021.3133691","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Transactions on Parallel and Distributed Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.8799999952316284,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[{"id":"https://openalex.org/G2827228560","display_name":null,"funder_award_id":"CCF-1812495","funder_id":"https://openalex.org/F4320337387","funder_display_name":"Division of Computing and Communication Foundations"},{"id":"https://openalex.org/G3312282444","display_name":null,"funder_award_id":"CCF-1901165","funder_id":"https://openalex.org/F4320337387","funder_display_name":"Division of Computing and Communication Foundations"},{"id":"https://openalex.org/G3932029878","display_name":null,"funder_award_id":"CCF-1702980","funder_id":"https://openalex.org/F4320337387","funder_display_name":"Division of Computing and Communication Foundations"},{"id":"https://openalex.org/G6194182566","display_name":null,"funder_award_id":"CCF-1703013","funder_id":"https://openalex.org/F4320337387","funder_display_name":"Division of Computing and Communication Foundations"},{"id":"https://openalex.org/G8908932306","display_name":null,"funder_award_id":"CCF-1936794","funder_id":"https://openalex.org/F4320337387","funder_display_name":"Division of Computing and Communication Foundations"}],"funders":[{"id":"https://openalex.org/F4320337387","display_name":"Division of Computing and Communication Foundations","ror":"https://ror.org/01mng8331"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":71,"referenced_works":["https://openalex.org/W637153065","https://openalex.org/W2010418442","https://openalex.org/W2030758178","https://openalex.org/W2068326054","https://openalex.org/W2094756095","https://openalex.org/W2282294254","https://openalex.org/W2284169075","https://openalex.org/W2285660444","https://openalex.org/W2442974303","https://openalex.org/W2515673159","https://openalex.org/W2516141709","https://openalex.org/W2583784797","https://openalex.org/W2584311934","https://openalex.org/W2606722458","https://openalex.org/W2625457103","https://openalex.org/W2762390662","https://openalex.org/W2765329037","https://openalex.org/W2794952988","https://openalex.org/W2795118915","https://openalex.org/W2883638665","https://openalex.org/W2902646240","https://openalex.org/W2926442184","https://openalex.org/W2951591319","https://openalex.org/W2952707032","https://openalex.org/W2954581201","https://openalex.org/W2957048678","https://openalex.org/W2962711740","https://openalex.org/W2962903741","https://openalex.org/W2964015378","https://openalex.org/W2964321699","https://openalex.org/W2964571482","https://openalex.org/W2979439447","https://openalex.org/W2979858238","https://openalex.org/W2980039165","https://openalex.org/W3000310304","https://openalex.org/W3004208721","https://openalex.org/W3011667710","https://openalex.org/W3012562343","https://openalex.org/W3016832937","https://openalex.org/W3017228913","https://openalex.org/W3034275286","https://openalex.org/W3034681945","https://openalex.org/W3043548994","https://openalex.org/W3047846843","https://openalex.org/W3091862797","https://openalex.org/W3092530991","https://openalex.org/W3096966254","https://openalex.org/W3102746402","https://openalex.org/W3105753905","https://openalex.org/W3105802176","https://openalex.org/W3106125969","https://openalex.org/W3123909522","https://openalex.org/W3157609068","https://openalex.org/W3158126894","https://openalex.org/W4210257598","https://openalex.org/W4237819810","https://openalex.org/W4240168186","https://openalex.org/W4241140669","https://openalex.org/W4294558607","https://openalex.org/W4302334595","https://openalex.org/W6620673361","https://openalex.org/W6720006811","https://openalex.org/W6726873649","https://openalex.org/W6738964360","https://openalex.org/W6754929296","https://openalex.org/W6760755035","https://openalex.org/W6781160529","https://openalex.org/W6785476653","https://openalex.org/W6789720248","https://openalex.org/W6794273881","https://openalex.org/W6807384801"],"related_works":["https://openalex.org/W2293118914","https://openalex.org/W2998381397","https://openalex.org/W4236419692","https://openalex.org/W2171015181","https://openalex.org/W4319952061","https://openalex.org/W4280636456","https://openalex.org/W4388913998","https://openalex.org/W4310584535","https://openalex.org/W4295935044","https://openalex.org/W3159906349"],"abstract_inverted_index":{"Graph":[0],"Convolutional":[1],"Neural":[2],"Networks":[3],"(GCNs)":[4],"have":[5],"emerged":[6],"as":[7],"promising":[8],"tools":[9],"for":[10,28,48,83],"graph-based":[11],"machine":[12],"learning":[13],"applications.":[14],"Given":[15],"that":[16,60,92,130,152],"GCNs":[17],"are":[18],"both":[19],"compute-":[20],"and":[21,51,86,138,157,160,170,176],"memory-intensive,":[22],"this":[23,38],"constitutes":[24],"a":[25,43,88,140,165],"major":[26],"challenge":[27],"the":[29,49,69,79,110,116,132,145],"underlying":[30],"hardware":[31],"to":[32,94,98,108,143,173],"efficiently":[33],"process":[34],"large-scale":[35],"GCNs.":[36,55,122],"In":[37],"paper,":[39],"we":[40,76],"introduce":[41],"SGCNAX,":[42],"scalable":[44],"GCN":[45,58,84,90,96],"accelerator":[46],"architecture":[47,129],"high-performance":[50],"energy-efficient":[52],"acceleration":[53,85],"of":[54,119,167],"Unlike":[56],"prior":[57],"accelerators":[59],"either":[61],"employ":[62],"limited":[63],"loop":[64,80],"optimization":[65,81],"techniques,":[66],"or":[67],"determine":[68],"design":[70],"variables":[71],"based":[72],"on":[73],"random":[74],"sampling,":[75],"systematically":[77],"explore":[78],"techniques":[82,107],"propose":[87,104],"flexible":[89],"dataflow":[91],"adapts":[93],"different":[95],"configurations":[97],"achieve":[99],"optimal":[100],"efficiency.":[101],"We":[102],"further":[103],"two":[105],"hardware-based":[106],"address":[109],"workload":[111,136,147],"imbalance":[112],"problem":[113],"caused":[114],"by":[115,164],"unbalanced":[117],"distribution":[118],"zeros":[120],"in":[121],"Specifically,":[123],"SGCNAX":[124,153],"exploits":[125],"an":[126],"outer-product-based":[127],"computation":[128],"mitigates":[131],"intra-PE":[133],"(Processing":[134],"Elements)":[135],"imbalance,":[137],"employs":[139],"group-and-shuffle":[141],"approach":[142],"mitigate":[144],"inter-PE":[146],"imbalance.":[148],"Simulation":[149],"results":[150],"show":[151],"performs":[154],"9.2x,":[155],"1.6x":[156],"1.2x":[158,171],"better,":[159],"reduces":[161],"DRAM":[162],"accesses":[163],"factor":[166],"9.7x,":[168],"2.9x":[169],"compared":[172],"HyGCN,":[174],"AWB-GCN,":[175],"GCNAX,":[177],"respectively.":[178]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":6},{"year":2023,"cited_by_count":8},{"year":2022,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}