{"id":"https://openalex.org/W4392648590","doi":"https://doi.org/10.1007/s11227-024-05949-6","title":"Block-wise dynamic mixed-precision for sparse matrix-vector multiplication on GPUs","display_name":"Block-wise dynamic mixed-precision for sparse matrix-vector multiplication on GPUs","publication_year":2024,"publication_date":"2024-03-11","ids":{"openalex":"https://openalex.org/W4392648590","doi":"https://doi.org/10.1007/s11227-024-05949-6"},"language":"en","primary_location":{"id":"doi:10.1007/s11227-024-05949-6","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s11227-024-05949-6","pdf_url":"https://link.springer.com/content/pdf/10.1007/s11227-024-05949-6.pdf","source":{"id":"https://openalex.org/S32326811","display_name":"The Journal of Supercomputing","issn_l":"0920-8542","issn":["0920-8542","1573-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Journal of Supercomputing","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s11227-024-05949-6.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069579553","display_name":"Zhixiang Zhao","orcid":"https://orcid.org/0000-0002-7893-2719"},"institutions":[{"id":"https://openalex.org/I151727225","display_name":"Harbin Engineering University","ror":"https://ror.org/03x80pn82","country_code":"CN","type":"education","lineage":["https://openalex.org/I151727225"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhixiang Zhao","raw_affiliation_strings":["College of Computer Science and Technology, Harbin Engineering University, Harbin, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Technology, Harbin Engineering University, Harbin, China","institution_ids":["https://openalex.org/I151727225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074415234","display_name":"Guoyin Zhang","orcid":"https://orcid.org/0000-0003-4833-1741"},"institutions":[{"id":"https://openalex.org/I151727225","display_name":"Harbin Engineering University","ror":"https://ror.org/03x80pn82","country_code":"CN","type":"education","lineage":["https://openalex.org/I151727225"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guoyin Zhang","raw_affiliation_strings":["College of Computer Science and Technology, Harbin Engineering University, Harbin, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Technology, Harbin Engineering University, Harbin, China","institution_ids":["https://openalex.org/I151727225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027083552","display_name":"Yanxia Wu","orcid":"https://orcid.org/0000-0001-8384-9234"},"institutions":[{"id":"https://openalex.org/I151727225","display_name":"Harbin Engineering University","ror":"https://ror.org/03x80pn82","country_code":"CN","type":"education","lineage":["https://openalex.org/I151727225"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanxia Wu","raw_affiliation_strings":["College of Computer Science and Technology, Harbin Engineering University, Harbin, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Technology, Harbin Engineering University, Harbin, China","institution_ids":["https://openalex.org/I151727225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100777951","display_name":"Ruize Hong","orcid":null},"institutions":[{"id":"https://openalex.org/I151727225","display_name":"Harbin Engineering University","ror":"https://ror.org/03x80pn82","country_code":"CN","type":"education","lineage":["https://openalex.org/I151727225"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ruize Hong","raw_affiliation_strings":["College of Computer Science and Technology, Harbin Engineering University, Harbin, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Technology, Harbin Engineering University, Harbin, China","institution_ids":["https://openalex.org/I151727225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025735969","display_name":"Yiqing Yang","orcid":"https://orcid.org/0000-0002-6522-6209"},"institutions":[{"id":"https://openalex.org/I151727225","display_name":"Harbin Engineering University","ror":"https://ror.org/03x80pn82","country_code":"CN","type":"education","lineage":["https://openalex.org/I151727225"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yiqing Yang","raw_affiliation_strings":["College of Computer Science and Technology, Harbin Engineering University, Harbin, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Technology, Harbin Engineering University, Harbin, China","institution_ids":["https://openalex.org/I151727225"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110082043","display_name":"Yan Fu","orcid":"https://orcid.org/0000-0001-6896-5931"},"institutions":[{"id":"https://openalex.org/I151727225","display_name":"Harbin Engineering University","ror":"https://ror.org/03x80pn82","country_code":"CN","type":"education","lineage":["https://openalex.org/I151727225"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yan Fu","raw_affiliation_strings":["College of Computer Science and Technology, Harbin Engineering University, Harbin, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Technology, Harbin Engineering University, Harbin, China","institution_ids":["https://openalex.org/I151727225"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5110082043"],"corresponding_institution_ids":["https://openalex.org/I151727225"],"apc_list":{"value":2390,"currency":"EUR","value_usd":2990},"apc_paid":{"value":2390,"currency":"EUR","value_usd":2990},"fwci":2.2021,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.87583809,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":"80","issue":"10","first_page":"13681","last_page":"13713"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9998000264167786,"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":0.9998000264167786,"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/T10792","display_name":"Matrix Theory and Algorithms","score":0.9972000122070312,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9937000274658203,"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":0.8381251096725464},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6491740942001343},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.595882773399353},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.5223763585090637},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.5100215077400208},{"id":"https://openalex.org/keywords/multiplication","display_name":"Multiplication (music)","score":0.48660552501678467},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.48624077439308167},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.4682745933532715},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4592895209789276},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.43197235465049744},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.4220307171344757},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.3364208936691284},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.10041928291320801},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.0936184823513031}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8381251096725464},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6491740942001343},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.595882773399353},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.5223763585090637},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.5100215077400208},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.48660552501678467},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.48624077439308167},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.4682745933532715},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4592895209789276},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.43197235465049744},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.4220307171344757},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3364208936691284},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.10041928291320801},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0936184823513031},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"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/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"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:10.1007/s11227-024-05949-6","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s11227-024-05949-6","pdf_url":"https://link.springer.com/content/pdf/10.1007/s11227-024-05949-6.pdf","source":{"id":"https://openalex.org/S32326811","display_name":"The Journal of Supercomputing","issn_l":"0920-8542","issn":["0920-8542","1573-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Journal of Supercomputing","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1007/s11227-024-05949-6","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s11227-024-05949-6","pdf_url":"https://link.springer.com/content/pdf/10.1007/s11227-024-05949-6.pdf","source":{"id":"https://openalex.org/S32326811","display_name":"The Journal of Supercomputing","issn_l":"0920-8542","issn":["0920-8542","1573-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Journal of Supercomputing","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4392648590.pdf"},"referenced_works_count":48,"referenced_works":["https://openalex.org/W794951912","https://openalex.org/W1588915715","https://openalex.org/W1965551736","https://openalex.org/W1972781306","https://openalex.org/W1988311079","https://openalex.org/W1996274599","https://openalex.org/W1997955449","https://openalex.org/W2007441708","https://openalex.org/W2034258707","https://openalex.org/W2035080386","https://openalex.org/W2043105874","https://openalex.org/W2059254721","https://openalex.org/W2082238959","https://openalex.org/W2091908315","https://openalex.org/W2095292158","https://openalex.org/W2101511474","https://openalex.org/W2102528350","https://openalex.org/W2128853364","https://openalex.org/W2162283062","https://openalex.org/W2339718050","https://openalex.org/W2568669383","https://openalex.org/W2570841452","https://openalex.org/W2741427986","https://openalex.org/W2766501636","https://openalex.org/W2781752755","https://openalex.org/W2887822915","https://openalex.org/W2909974497","https://openalex.org/W2996522450","https://openalex.org/W3009762362","https://openalex.org/W3011554625","https://openalex.org/W3091170309","https://openalex.org/W3093098265","https://openalex.org/W3129303295","https://openalex.org/W3137131725","https://openalex.org/W3138530731","https://openalex.org/W3175189837","https://openalex.org/W3185108972","https://openalex.org/W3186231168","https://openalex.org/W3196504174","https://openalex.org/W4214570494","https://openalex.org/W4226207577","https://openalex.org/W4243261006","https://openalex.org/W4247828381","https://openalex.org/W4281657881","https://openalex.org/W4283270874","https://openalex.org/W4285104934","https://openalex.org/W4308524155","https://openalex.org/W4312068769"],"related_works":["https://openalex.org/W3099313426","https://openalex.org/W4287593139","https://openalex.org/W752783541","https://openalex.org/W1506547947","https://openalex.org/W2995605830","https://openalex.org/W4206811032","https://openalex.org/W2086123442","https://openalex.org/W2596457687","https://openalex.org/W3212757063","https://openalex.org/W2093666864"],"abstract_inverted_index":{"Abstract":[0],"Sparse":[1],"matrix-vector":[2],"multiplication":[3],"(SpMV)":[4],"plays":[5],"a":[6,10,52,96],"critical":[7],"role":[8],"in":[9,18],"wide":[11],"range":[12],"of":[13,37,43,107,200,261],"linear":[14],"algebra":[15],"computations,":[16],"particularly":[17],"scientific":[19],"and":[20,35,84,124,128,162,172,185,193,206,220,232,242],"engineering":[21],"disciplines.":[22],"However,":[23],"the":[24,40,68,104,108,133,198,259,266],"irregular":[25],"memory":[26,30,163],"access":[27],"patterns,":[28],"extensive":[29],"usage,":[31],"high":[32],"bandwidth":[33],"requirements,":[34],"underutilization":[36],"parallelism":[38,161],"hinder":[39],"computational":[41],"efficiency":[42],"SpMV":[44,116,188],"on":[45,143,157,228,250],"GPUs.":[46],"In":[47],"this":[48],"paper,":[49],"we":[50,88],"propose":[51],"novel":[53],"approach":[54],"called":[55],"block-wise":[56],"dynamic":[57],"mixed-precision":[58],"(BDMP)":[59],"to":[60,92,212,234],"address":[61],"these":[62],"challenges.":[63],"Our":[64],"methodology":[65],"involves":[66],"partitioning":[67],"original":[69,109],"matrix":[70,134],"into":[71,102,135],"uniformly":[72],"sized":[73],"blocks,":[74],"with":[75,150,181],"each":[76,93,176],"block\u2019s":[77],"size":[78],"determined":[79],"by":[80,165],"considering":[81],"architectural":[82],"characteristics":[83],"accuracy":[85],"requirements.":[86],"Additionally,":[87],"dynamically":[89],"assign":[90],"precision":[91,97,267],"block":[94,144],"using":[95,195],"selection":[98],"method":[99],"that":[100,256],"takes":[101],"account":[103],"value":[105],"distribution":[106],"sparse":[110],"matrix.":[111],"We":[112,178],"develop":[113],"two":[114,136],"distinct":[115],"computation":[117,262],"algorithms":[118],"for":[119,139,148,175,269],"BDMP:":[120],"BDMP-PBP":[121,131,205],"(Precision-based":[122],"partitioning)":[123],"BDMP-TCKI":[125,155,207],"(Tailored":[126],"compression":[127],"kernel":[129,173],"implementation).":[130],"partitions":[132],"independent":[137],"matrices":[138,196],"separate":[140],"computations":[141],"based":[142],"precision,":[145],"offering":[146],"flexibility":[147],"integration":[149],"other":[151],"optimization":[152,260],"techniques.":[153],"Meanwhile,":[154],"focuses":[156],"achieving":[158],"significant":[159],"thread-level":[160],"utilization":[164],"tailoring":[166],"an":[167],"appropriate":[168],"compressed":[169],"storage":[170],"format":[171],"implementation":[174],"block.":[177],"compare":[179],"BDMP":[180,257],"NVIDIA\u2019s":[182],"cuSPARSE":[183],"library":[184],"three":[186],"state-of-the-art":[187],"methods,":[189],"including":[190],"SELLP,":[191],"MergeBase,":[192],"BalanceCSR,":[194],"from":[197],"University":[199],"Florida\u2019s":[201],"SuiteSparse":[202],"dataset":[203],"collection.":[204],"show":[208],"average":[209],"speedups":[210],"up":[211,233],"2.64":[213],"$$\\times":[214,222,236,244],"$$":[215,223,237,245],"<mml:math":[216,224,238,246],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\">":[217,225,239,247],"<mml:mo>\u00d7</mml:mo>":[218,226,240,248],"</mml:math>":[219,227,241,249],"2.91":[221],"Turing":[229],"RTX":[230],"2080Ti,":[231],"2.99":[235],"3.22":[243],"Ampere":[251],"A100.":[252],"The":[253],"results":[254],"demonstrate":[255],"enables":[258],"speed":[263],"without":[264],"compromising":[265],"necessary":[268],"reliable":[270],"results.":[271]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4}],"updated_date":"2026-06-13T06:13:01.061226","created_date":"2025-10-10T00:00:00"}
