{"id":"https://openalex.org/W4391986945","doi":"https://doi.org/10.1145/3627535.3638470","title":"A Row Decomposition-based Approach for Sparse Matrix Multiplication on GPUs","display_name":"A Row Decomposition-based Approach for Sparse Matrix Multiplication on GPUs","publication_year":2024,"publication_date":"2024-02-20","ids":{"openalex":"https://openalex.org/W4391986945","doi":"https://doi.org/10.1145/3627535.3638470"},"language":"en","primary_location":{"id":"doi:10.1145/3627535.3638470","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3627535.3638470","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3627535.3638470","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 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3627535.3638470","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5028520638","display_name":"Meng Pang","orcid":"https://orcid.org/0009-0005-5710-0285"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Meng Pang","raw_affiliation_strings":["Department of Computer Science and Technology, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0005-5710-0285","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Technology, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100660008","display_name":"Xiang Fei","orcid":"https://orcid.org/0000-0002-9549-612X"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiang Fei","raw_affiliation_strings":["Department of Computer Science and Technology, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-9549-612X","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Technology, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042878427","display_name":"Peng Qu","orcid":"https://orcid.org/0000-0002-1786-5372"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Peng Qu","raw_affiliation_strings":["Department of Computer Science and Technology, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-1786-5372","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Technology, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016596981","display_name":"Youhui Zhang","orcid":"https://orcid.org/0000-0003-2333-3580"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Youhui Zhang","raw_affiliation_strings":["Department of Computer Science and Technology, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-2333-3580","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Technology, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5057648062","display_name":"Zhaolin Li","orcid":"https://orcid.org/0009-0001-1131-3024"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhaolin Li","raw_affiliation_strings":["Department of Computer Science and Technology, Tsinghua University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0001-1131-3024","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Technology, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":23,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"377","last_page":"389"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9984999895095825,"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":0.9984999895095825,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9940999746322632,"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"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9865999817848206,"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/speedup","display_name":"Speedup","score":0.8143438100814819},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.7915431261062622},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7292272448539734},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.696757435798645},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.626953125},{"id":"https://openalex.org/keywords/multiplication","display_name":"Multiplication (music)","score":0.5623490214347839},{"id":"https://openalex.org/keywords/lu-decomposition","display_name":"LU decomposition","score":0.5520315170288086},{"id":"https://openalex.org/keywords/matrix-decomposition","display_name":"Matrix decomposition","score":0.48565447330474854},{"id":"https://openalex.org/keywords/preprocessor","display_name":"Preprocessor","score":0.4730895161628723},{"id":"https://openalex.org/keywords/row","display_name":"Row","score":0.47034838795661926},{"id":"https://openalex.org/keywords/strassen-algorithm","display_name":"Strassen algorithm","score":0.46634620428085327},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4616272747516632},{"id":"https://openalex.org/keywords/matrix","display_name":"Matrix (chemical analysis)","score":0.4334467351436615},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.42339959740638733},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.4122239351272583},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.17665424942970276},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.17238110303878784}],"concepts":[{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8143438100814819},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.7915431261062622},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7292272448539734},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.696757435798645},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.626953125},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.5623490214347839},{"id":"https://openalex.org/C123213974","wikidata":"https://www.wikidata.org/wiki/Q833089","display_name":"LU decomposition","level":4,"score":0.5520315170288086},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.48565447330474854},{"id":"https://openalex.org/C34736171","wikidata":"https://www.wikidata.org/wiki/Q918333","display_name":"Preprocessor","level":2,"score":0.4730895161628723},{"id":"https://openalex.org/C135598885","wikidata":"https://www.wikidata.org/wiki/Q1366302","display_name":"Row","level":2,"score":0.47034838795661926},{"id":"https://openalex.org/C39096654","wikidata":"https://www.wikidata.org/wiki/Q728507","display_name":"Strassen algorithm","level":4,"score":0.46634620428085327},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4616272747516632},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.4334467351436615},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.42339959740638733},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4122239351272583},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.17665424942970276},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.17238110303878784},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.0},{"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/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"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/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","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},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3627535.3638470","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3627535.3638470","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3627535.3638470","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 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3627535.3638470","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3627535.3638470","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3627535.3638470","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 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3911511746","display_name":null,"funder_award_id":"62250006","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6132722138","display_name":"\u7c7b\u8111\u8ba1\u7b97\u7cfb\u7edf\u5c42\u6b21\u7ed3\u6784\u4e0e\u5b8c\u5907\u6027\u7814\u7a76","funder_award_id":"62072266","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G835360120","display_name":null,"funder_award_id":"BNR2022RC01003","funder_id":"https://openalex.org/F4320329777","funder_display_name":"Beijing National Research Center For Information Science And Technology"},{"id":"https://openalex.org/G8764175072","display_name":null,"funder_award_id":"62202254","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322392","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549"},{"id":"https://openalex.org/F4320329777","display_name":"Beijing National Research Center For Information Science And Technology","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4391986945.pdf","grobid_xml":"https://content.openalex.org/works/W4391986945.grobid-xml"},"referenced_works_count":22,"referenced_works":["https://openalex.org/W1985312666","https://openalex.org/W1990832096","https://openalex.org/W1999702719","https://openalex.org/W2035080386","https://openalex.org/W2040404421","https://openalex.org/W2054141820","https://openalex.org/W2099625934","https://openalex.org/W2126917700","https://openalex.org/W2249508874","https://openalex.org/W2590246587","https://openalex.org/W2617403939","https://openalex.org/W2626696598","https://openalex.org/W2808128431","https://openalex.org/W2914631005","https://openalex.org/W2964337156","https://openalex.org/W2999347085","https://openalex.org/W3108012228","https://openalex.org/W3208099998","https://openalex.org/W4231510805","https://openalex.org/W4234552385","https://openalex.org/W4310563332","https://openalex.org/W4379518528"],"related_works":["https://openalex.org/W1974407206","https://openalex.org/W1829171354","https://openalex.org/W2004916460","https://openalex.org/W2962985479","https://openalex.org/W1523285705","https://openalex.org/W1972907356","https://openalex.org/W4289143641","https://openalex.org/W2901988615","https://openalex.org/W3091817928","https://openalex.org/W4386764095"],"abstract_inverted_index":{"Sparse-Matrix":[0],"Dense-Matrix":[1],"Multiplication":[2,9],"(SpMM)":[3],"and":[4,28,35,87,103,118,146],"Sampled":[5],"Dense":[6,7],"Matrix":[8],"(SDDMM)":[10],"are":[11],"important":[12],"sparse":[13,26,34,44,81],"kernels":[14,66],"in":[15,24,170],"various":[16],"computation":[17],"domains.":[18],"The":[19],"uneven":[20],"distribution":[21],"of":[22,134,142,149,157,185],"nonzeros":[23],"the":[25,29,52,64,70,80,99,126,161,171,186],"matrix":[27,45,82],"tight":[30],"data":[31],"dependence":[32],"between":[33],"dense":[36],"matrices":[37],"make":[38],"it":[39],"a":[40,57,132,139,147,154],"challenge":[41],"to":[42,62,90,125,136,151],"run":[43],"multiplication":[46],"efficiently":[47],"on":[48,67],"GPUs.":[49],"By":[50],"analyzing":[51],"aforementioned":[53],"problems,":[54],"we":[55],"propose":[56],"row":[58],"decomposition":[59],"(RoDe)-based":[60],"approach":[61],"optimize":[63,92],"two":[65],"GPUs,":[68],"using":[69],"standard":[71],"Compressed":[72],"Sparse":[73],"Row":[74],"(CSR)":[75],"format.":[76],"Specifically,":[77],"RoDe":[78,111,130,165],"divides":[79],"rows":[83],"into":[84],"regular":[85],"parts":[86],"residual":[88],"parts,":[89],"fully":[91],"their":[93],"computations":[94],"separately.":[95],"We":[96],"also":[97,166],"devise":[98],"corresponding":[100],"load":[101],"balancing":[102],"finegrained":[104],"pipelining":[105],"technologies.":[106],"Profiling":[107],"results":[108],"show":[109],"that":[110],"can":[112],"achieve":[113],"more":[114],"efficient":[115],"memory":[116],"access":[117],"reduce":[119],"warp":[120],"stall":[121],"cycles":[122],"significantly.":[123],"Compared":[124],"state-of-the-art":[127],"(SOTA)":[128],"alternatives,":[129],"achieves":[131],"speedup":[133,148],"up":[135,150],"7.86\u00d7":[137],"with":[138,153],"geometric":[140,155],"mean":[141,156],"1.45\u00d7":[143],"for":[144,159],"SpMM,":[145],"8.99\u00d7":[152],"1.49\u00d7":[158],"SDDMM;":[160],"dataset":[162],"is":[163,179],"SuiteSparse.":[164],"outperforms":[167],"its":[168,176],"counterpart":[169],"deep":[172],"learning":[173],"dataset.":[174],"Furthermore,":[175],"preprocessing":[177],"overhead":[178],"significantly":[180],"smaller,":[181],"averaging":[182],"only":[183],"16%":[184],"SOTA.":[187]},"counts_by_year":[{"year":2026,"cited_by_count":9},{"year":2025,"cited_by_count":13},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}