{"id":"https://openalex.org/W4408029514","doi":"https://doi.org/10.1145/3710848.3710858","title":"FlashSparse: Minimizing Computation Redundancy for Fast Sparse Matrix Multiplications on Tensor Cores","display_name":"FlashSparse: Minimizing Computation Redundancy for Fast Sparse Matrix Multiplications on Tensor Cores","publication_year":2025,"publication_date":"2025-02-28","ids":{"openalex":"https://openalex.org/W4408029514","doi":"https://doi.org/10.1145/3710848.3710858"},"language":"en","primary_location":{"id":"doi:10.1145/3710848.3710858","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3710848.3710858","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3710848.3710858","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3710848.3710858","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100837458","display_name":"Jinliang Shi","orcid":"https://orcid.org/0000-0001-5420-906X"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinliang Shi","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-5420-906X","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088172355","display_name":"Shigang Li","orcid":"https://orcid.org/0000-0003-0022-7865"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shigang Li","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-0022-7865","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101067362","display_name":"Yingcheng Xu","orcid":"https://orcid.org/0009-0007-1698-3565"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Youxuan Xu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0007-1698-3565","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065684265","display_name":"Rongtian Fu","orcid":"https://orcid.org/0009-0008-7636-3082"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rongtian Fu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0008-7636-3082","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100425338","display_name":"Xueying Wang","orcid":"https://orcid.org/0000-0002-7835-113X"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xueying Wang","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-7835-113X","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5112858390","display_name":"Tong Wu","orcid":"https://orcid.org/0009-0001-0472-5178"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tong Wu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0001-0472-5178","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I139759216"],"apc_list":null,"apc_paid":null,"fwci":14.204,"has_fulltext":false,"cited_by_count":10,"citation_normalized_percentile":{"value":0.99132631,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"312","last_page":"325"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9991999864578247,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9991999864578247,"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"}},{"id":"https://openalex.org/T12303","display_name":"Tensor decomposition and applications","score":0.996399998664856,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9848999977111816,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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":0.6453133225440979},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.6333938837051392},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5997381210327148},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.5833938717842102},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5590627193450928},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.5097209811210632},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.4430708885192871},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.419596403837204},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.27679115533828735},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.07690006494522095}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6453133225440979},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.6333938837051392},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5997381210327148},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.5833938717842102},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5590627193450928},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.5097209811210632},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.4430708885192871},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.419596403837204},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.27679115533828735},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.07690006494522095},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","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},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3710848.3710858","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3710848.3710858","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3710848.3710858","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3710848.3710858","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3710848.3710858","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3710848.3710858","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1764641568","display_name":null,"funder_award_id":"2023ZD0120502","funder_id":"https://openalex.org/F4320323817","funder_display_name":"Universitas Brawijaya"},{"id":"https://openalex.org/G3475808386","display_name":null,"funder_award_id":"62372055","funder_id":"https://openalex.org/F4320323817","funder_display_name":"Universitas Brawijaya"}],"funders":[{"id":"https://openalex.org/F4320323817","display_name":"Universitas Brawijaya","ror":"https://ror.org/01wk3d929"}],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4408029514.pdf"},"referenced_works_count":27,"referenced_works":["https://openalex.org/W2002096154","https://openalex.org/W2035080386","https://openalex.org/W2123736933","https://openalex.org/W2126917700","https://openalex.org/W2136583165","https://openalex.org/W2161902954","https://openalex.org/W2734941459","https://openalex.org/W2901549770","https://openalex.org/W2914631005","https://openalex.org/W2999347085","https://openalex.org/W3006432778","https://openalex.org/W3104528661","https://openalex.org/W3123909522","https://openalex.org/W3132695675","https://openalex.org/W3175925542","https://openalex.org/W3208099998","https://openalex.org/W4214940715","https://openalex.org/W4220912491","https://openalex.org/W4226283934","https://openalex.org/W4231510805","https://openalex.org/W4283462727","https://openalex.org/W4310563332","https://openalex.org/W4366198975","https://openalex.org/W4381832157","https://openalex.org/W4388661983","https://openalex.org/W4391986945","https://openalex.org/W4394998407"],"related_works":["https://openalex.org/W1925544630","https://openalex.org/W2293771254","https://openalex.org/W2004686618","https://openalex.org/W3202552726","https://openalex.org/W2039875226","https://openalex.org/W4221142455","https://openalex.org/W3121828480","https://openalex.org/W2914631005","https://openalex.org/W2123154672","https://openalex.org/W2032786851"],"abstract_inverted_index":{"Sparse":[0],"Matrix-matrix":[1],"Multiplication":[2,8],"(SpMM)":[3],"and":[4,17,79,91,137,152,173],"Sampled":[5],"Dense-dense":[6],"Matrix":[7],"(SDDMM)":[9],"are":[10],"important":[11],"sparse":[12,52,77,87,106,139,164],"operators":[13,40],"in":[14],"scientific":[15],"computing":[16,29,116],"deep":[18],"learning.":[19],"Tensor":[20],"Core":[21],"Units":[22],"(TCUs)":[23],"enhance":[24],"modern":[25],"accelerators":[26],"with":[27,127],"superior":[28],"power,":[30],"which":[31],"is":[32,55,111,120,125],"promising":[33],"to":[34,41,47,57,72,143],"boost":[35],"the":[36,48,74,80,86,104,108,115],"performance":[37],"of":[38,50,118],"matrix":[39,99,140,165],"a":[42,69,96,128,138,160],"higher":[43],"level.":[44],"However,":[45],"due":[46],"irregularity":[49],"unstructured":[51],"data,":[53],"it":[54],"difficult":[56],"deliver":[58],"practical":[59],"speedups":[60],"on":[61,93,150],"TCUs.":[62],"To":[63],"this":[64],"end,":[65],"we":[66],"propose":[67],"FlashSparse,":[68],"novel":[70,97],"approach":[71],"bridge":[73],"gap":[75],"between":[76],"workloads":[78],"TCU":[81],"architecture.":[82],"Specifically,":[83],"FlashSparse":[84,124,158],"minimizes":[85],"granularity":[88],"for":[89,133,163],"SpMM":[90],"SDDMM":[92],"TCUs":[94,119],"through":[95],"swap-and-transpose":[98],"multiplication":[100],"strategy.":[101],"Benefiting":[102],"from":[103],"minimum":[105],"granularity,":[107],"computation":[109],"redundancy":[110],"remarkably":[112],"reduced":[113],"while":[114],"power":[117],"fully":[121],"utilized.":[122],"Besides,":[123],"equipped":[126],"memory-efficient":[129],"thread":[130],"mapping":[131],"strategy":[132],"coalesced":[134],"data":[135],"access":[136],"storage":[141],"format":[142],"save":[144],"memory":[145],"footprint.":[146],"Extensive":[147],"experimental":[148],"results":[149],"H100":[151],"RTX":[153],"4090":[154],"GPUs":[155],"show":[156],"that":[157],"sets":[159],"new":[161],"state-of-the-art":[162],"multiplications":[166],"(geometric":[167],"mean":[168],"5.5x":[169],"speedup":[170,175],"over":[171,176],"DTC-SpMM":[172],"3.22x":[174],"RoDe).":[177]},"counts_by_year":[{"year":2026,"cited_by_count":8},{"year":2025,"cited_by_count":2}],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-10T00:00:00"}