{"id":"https://openalex.org/W4321637181","doi":"https://doi.org/10.1109/fpl57034.2022.00054","title":"SDMA: An Efficient and Flexible Sparse-Dense Matrix-Multiplication Architecture for GNNs","display_name":"SDMA: An Efficient and Flexible Sparse-Dense Matrix-Multiplication Architecture for GNNs","publication_year":2022,"publication_date":"2022-08-01","ids":{"openalex":"https://openalex.org/W4321637181","doi":"https://doi.org/10.1109/fpl57034.2022.00054"},"language":"en","primary_location":{"id":"doi:10.1109/fpl57034.2022.00054","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/fpl57034.2022.00054","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 32nd International Conference on Field-Programmable Logic and Applications (FPL)","raw_type":"proceedings-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/A5061329069","display_name":"Yingxue Gao","orcid":"https://orcid.org/0000-0002-0833-1379"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yingxue Gao","raw_affiliation_strings":["University Of Science and Technology Of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University Of Science and Technology Of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101642099","display_name":"Lei Gong","orcid":"https://orcid.org/0000-0001-6017-8490"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lei Gong","raw_affiliation_strings":["University Of Science and Technology Of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University Of Science and Technology Of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100406973","display_name":"Chao Wang","orcid":"https://orcid.org/0000-0002-3262-653X"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chao Wang","raw_affiliation_strings":["University Of Science and Technology Of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University Of Science and Technology Of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089833377","display_name":"Teng Wang","orcid":"https://orcid.org/0000-0002-7281-1203"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Teng Wang","raw_affiliation_strings":["University Of Science and Technology Of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University Of Science and Technology Of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077322091","display_name":"Xuehai Zhou","orcid":"https://orcid.org/0000-0002-8360-3143"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuehai Zhou","raw_affiliation_strings":["University Of Science and Technology Of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University Of Science and Technology Of China","institution_ids":["https://openalex.org/I126520041"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I126520041"],"apc_list":null,"apc_paid":null,"fwci":1.1102,"has_fulltext":false,"cited_by_count":8,"citation_normalized_percentile":{"value":0.82125258,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":"abs/2112.13389","issue":null,"first_page":"307","last_page":"312"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9991000294685364,"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":0.9991000294685364,"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":0.9926000237464905,"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/T12326","display_name":"Network Packet Processing and Optimization","score":0.9912999868392944,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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":0.8377429246902466},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5983878374099731},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.5066404938697815},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.48727476596832275},{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.4180670976638794},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.4126186966896057},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.33683258295059204},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.24680855870246887},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.20337620377540588},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.19552940130233765}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8377429246902466},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5983878374099731},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.5066404938697815},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.48727476596832275},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.4180670976638794},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.4126186966896057},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.33683258295059204},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.24680855870246887},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.20337620377540588},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.19552940130233765},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/fpl57034.2022.00054","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/fpl57034.2022.00054","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 32nd International Conference on Field-Programmable Logic and Applications (FPL)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.8999999761581421,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":45,"referenced_works":["https://openalex.org/W2016311778","https://openalex.org/W2067247412","https://openalex.org/W2152839228","https://openalex.org/W2406293284","https://openalex.org/W2529031466","https://openalex.org/W2584785188","https://openalex.org/W2807021761","https://openalex.org/W2883927138","https://openalex.org/W2891691622","https://openalex.org/W2903262661","https://openalex.org/W2933438941","https://openalex.org/W2964337156","https://openalex.org/W2964571482","https://openalex.org/W2981088532","https://openalex.org/W2984625009","https://openalex.org/W2999301586","https://openalex.org/W3017228913","https://openalex.org/W3026844281","https://openalex.org/W3043406639","https://openalex.org/W3047846843","https://openalex.org/W3084616068","https://openalex.org/W3100848837","https://openalex.org/W3103168911","https://openalex.org/W3103478730","https://openalex.org/W3105753905","https://openalex.org/W3130660608","https://openalex.org/W3132695675","https://openalex.org/W3132871189","https://openalex.org/W3137198614","https://openalex.org/W3154688839","https://openalex.org/W3155922894","https://openalex.org/W3157609068","https://openalex.org/W3158540161","https://openalex.org/W3159314463","https://openalex.org/W3170577894","https://openalex.org/W3191648064","https://openalex.org/W3198975860","https://openalex.org/W3214431650","https://openalex.org/W4226252440","https://openalex.org/W4231846589","https://openalex.org/W4288419263","https://openalex.org/W6603201521","https://openalex.org/W6681885597","https://openalex.org/W6796363179","https://openalex.org/W6811027350"],"related_works":["https://openalex.org/W2111241003","https://openalex.org/W2355315220","https://openalex.org/W4200391368","https://openalex.org/W2210979487","https://openalex.org/W2316202402","https://openalex.org/W2074043759","https://openalex.org/W2082487009","https://openalex.org/W2373535795","https://openalex.org/W1925544630","https://openalex.org/W2004686618"],"abstract_inverted_index":{"In":[0,58],"recent":[1],"years,":[2],"graph":[3,52],"neural":[4],"networks":[5],"(GNNs)":[6],"as":[7],"a":[8,146],"deep":[9],"learning":[10],"model":[11],"have":[12],"emerged.":[13],"Sparse-Dense":[14],"Matrix":[15],"Multiplication":[16],"(SpMM)":[17],"is":[18,125],"the":[19,36,51,55,76,99,139,154,169,177],"critical":[20],"component":[21],"of":[22,38,50,78,131],"GNNs.":[23,71],"However,":[24],"SpMM":[25,69],"involves":[26],"many":[27],"irregular":[28,82],"calculations":[29],"and":[30,41,47,81,137,172],"random":[31],"memory":[32,83],"accesses,":[33],"resulting":[34],"in":[35],"inefficiency":[37],"general-purpose":[39],"processors":[40],"dedicated":[42],"accelerators.":[43],"The":[44,93,109,159],"highly":[45],"sparse":[46,100],"uneven":[48],"distribution":[49],"further":[53],"exacerbates":[54],"above":[56,140],"problems.":[57],"this":[59],"work,":[60],"we":[61,135,150],"propose":[62],"SDMA,":[63],"an":[64],"efficient":[65],"architecture":[66],"to":[67,102,127,144],"accelerate":[68],"for":[70],"SDMA":[72,143,152,163],"can":[73,113],"collaboratively":[74],"address":[75],"challenges":[77],"load":[79,104],"imbalance":[80],"accesses.":[84],"We":[85],"first":[86],"present":[87],"three":[88],"hardware-oriented":[89],"optimization":[90,111,141],"methods:":[91],"1)":[92],"Equal-value":[94],"partition":[95],"method":[96,112,124],"effectively":[97],"divides":[98],"matrix":[101],"achieve":[103,145],"balancing":[105],"between":[106],"tiles.":[107],"2)":[108],"vertex-clustering":[110],"explore":[114],"more":[115],"data":[116],"locality.":[117],"3)":[118],"An":[119],"adaptive":[120],"on-chip":[121],"dataflow":[122],"scheduling":[123],"proposed":[126],"make":[128],"full":[129],"use":[130],"computing":[132],"resources.":[133],"Then,":[134],"combine":[136],"integrate":[138],"into":[142],"high-performance":[147],"architecture.":[148],"Finally,":[149],"prototype":[151],"on":[153],"Xilinx":[155],"Alveo":[156],"U50":[157],"FPGA.":[158],"results":[160],"demonstrate":[161],"that":[162],"achieves":[164],"2.19x-3.35x":[165],"energy":[166],"efficiency":[167,175],"over":[168,176],"GPU":[170],"implementation":[171],"2.03x":[173],"DSP":[174],"FPGA":[178],"implementation.":[179]},"counts_by_year":[{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":4}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}