{"id":"https://openalex.org/W3202517391","doi":"https://doi.org/10.1145/3472456.3472478","title":"Fast and Scalable Sparse Triangular Solver for Multi-GPU Based HPC Architectures","display_name":"Fast and Scalable Sparse Triangular Solver for Multi-GPU Based HPC Architectures","publication_year":2021,"publication_date":"2021-08-09","ids":{"openalex":"https://openalex.org/W3202517391","doi":"https://doi.org/10.1145/3472456.3472478","mag":"3202517391"},"language":"en","primary_location":{"id":"doi:10.1145/3472456.3472478","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3472456.3472478","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"50th International Conference on Parallel Processing","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/A5024508481","display_name":"Chenhao Xie","orcid":"https://orcid.org/0000-0002-1399-0352"},"institutions":[{"id":"https://openalex.org/I142606810","display_name":"Pacific Northwest National Laboratory","ror":"https://ror.org/05h992307","country_code":"US","type":"facility","lineage":["https://openalex.org/I1325736334","https://openalex.org/I1330989302","https://openalex.org/I142606810","https://openalex.org/I39565521"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"CHENHAO XIE","raw_affiliation_strings":["Pacific Northwest National Laboratory, United States of America"],"affiliations":[{"raw_affiliation_string":"Pacific Northwest National Laboratory, United States of America","institution_ids":["https://openalex.org/I142606810"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063649910","display_name":"Jieyang Chen","orcid":"https://orcid.org/0000-0002-1905-9171"},"institutions":[{"id":"https://openalex.org/I1289243028","display_name":"Oak Ridge National Laboratory","ror":"https://ror.org/01qz5mb56","country_code":"US","type":"facility","lineage":["https://openalex.org/I1289243028","https://openalex.org/I1330989302","https://openalex.org/I39565521","https://openalex.org/I4210159294"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jieyang Chen","raw_affiliation_strings":["Oak Ridge National Laboratory, United States of America"],"affiliations":[{"raw_affiliation_string":"Oak Ridge National Laboratory, United States of America","institution_ids":["https://openalex.org/I1289243028"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063476184","display_name":"Jesun Firoz","orcid":"https://orcid.org/0000-0002-8174-2545"},"institutions":[{"id":"https://openalex.org/I142606810","display_name":"Pacific Northwest National Laboratory","ror":"https://ror.org/05h992307","country_code":"US","type":"facility","lineage":["https://openalex.org/I1325736334","https://openalex.org/I1330989302","https://openalex.org/I142606810","https://openalex.org/I39565521"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jesun Firoz","raw_affiliation_strings":["Pacific Northwest National Laboratory, United States of America"],"affiliations":[{"raw_affiliation_string":"Pacific Northwest National Laboratory, United States of America","institution_ids":["https://openalex.org/I142606810"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100454305","display_name":"Jiajia Li","orcid":"https://orcid.org/0000-0003-1270-4147"},"institutions":[{"id":"https://openalex.org/I142606810","display_name":"Pacific Northwest National Laboratory","ror":"https://ror.org/05h992307","country_code":"US","type":"facility","lineage":["https://openalex.org/I1325736334","https://openalex.org/I1330989302","https://openalex.org/I142606810","https://openalex.org/I39565521"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiajia Li","raw_affiliation_strings":["Pacific Northwest National Laboratory, United States of America"],"affiliations":[{"raw_affiliation_string":"Pacific Northwest National Laboratory, United States of America","institution_ids":["https://openalex.org/I142606810"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043209884","display_name":"Shuaiwen Leon Song","orcid":"https://orcid.org/0000-0002-8402-1436"},"institutions":[{"id":"https://openalex.org/I129604602","display_name":"University of Sydney","ror":"https://ror.org/0384j8v12","country_code":"AU","type":"education","lineage":["https://openalex.org/I129604602"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Shuaiwen Leon Song","raw_affiliation_strings":["University of Sydney, Australia"],"affiliations":[{"raw_affiliation_string":"University of Sydney, Australia","institution_ids":["https://openalex.org/I129604602"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038062469","display_name":"Kevin Barker","orcid":"https://orcid.org/0000-0003-4947-0559"},"institutions":[{"id":"https://openalex.org/I142606810","display_name":"Pacific Northwest National Laboratory","ror":"https://ror.org/05h992307","country_code":"US","type":"facility","lineage":["https://openalex.org/I1325736334","https://openalex.org/I1330989302","https://openalex.org/I142606810","https://openalex.org/I39565521"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kevin Barker","raw_affiliation_strings":["Pacific Northwest National Laboratory, United States of America"],"affiliations":[{"raw_affiliation_string":"Pacific Northwest National Laboratory, United States of America","institution_ids":["https://openalex.org/I142606810"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016828976","display_name":"Mark Raugas","orcid":"https://orcid.org/0000-0002-5597-8781"},"institutions":[{"id":"https://openalex.org/I142606810","display_name":"Pacific Northwest National Laboratory","ror":"https://ror.org/05h992307","country_code":"US","type":"facility","lineage":["https://openalex.org/I1325736334","https://openalex.org/I1330989302","https://openalex.org/I142606810","https://openalex.org/I39565521"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mark Raugas","raw_affiliation_strings":["Pacific Northwest National Laboratory, United States of America"],"affiliations":[{"raw_affiliation_string":"Pacific Northwest National Laboratory, United States of America","institution_ids":["https://openalex.org/I142606810"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100413657","display_name":"Ang Li","orcid":"https://orcid.org/0000-0003-3734-9137"},"institutions":[{"id":"https://openalex.org/I142606810","display_name":"Pacific Northwest National Laboratory","ror":"https://ror.org/05h992307","country_code":"US","type":"facility","lineage":["https://openalex.org/I1325736334","https://openalex.org/I1330989302","https://openalex.org/I142606810","https://openalex.org/I39565521"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ang Li","raw_affiliation_strings":["Pacific Northwest National Laboratory, United States of America"],"affiliations":[{"raw_affiliation_string":"Pacific Northwest National Laboratory, United States of America","institution_ids":["https://openalex.org/I142606810"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5024508481"],"corresponding_institution_ids":["https://openalex.org/I142606810"],"apc_list":null,"apc_paid":null,"fwci":1.4058,"has_fulltext":false,"cited_by_count":10,"citation_normalized_percentile":{"value":0.80787645,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"11"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":1.0,"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":1.0,"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/T10829","display_name":"Interconnection Networks and Systems","score":0.9991000294685364,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9979000091552734,"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":0.8877536058425903},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.7794649600982666},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7399734854698181},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.7314894199371338},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.5406578779220581},{"id":"https://openalex.org/keywords/synchronization","display_name":"Synchronization (alternating current)","score":0.5022478103637695},{"id":"https://openalex.org/keywords/solver","display_name":"Solver","score":0.49039149284362793},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.44679901003837585},{"id":"https://openalex.org/keywords/general-purpose-computing-on-graphics-processing-units","display_name":"General-purpose computing on graphics processing units","score":0.4316745102405548},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.4281182885169983},{"id":"https://openalex.org/keywords/programming-paradigm","display_name":"Programming paradigm","score":0.4201643764972687},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.34611210227012634},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.1894223690032959},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.1502009630203247}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8877536058425903},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.7794649600982666},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7399734854698181},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7314894199371338},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5406578779220581},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.5022478103637695},{"id":"https://openalex.org/C2778770139","wikidata":"https://www.wikidata.org/wiki/Q1966904","display_name":"Solver","level":2,"score":0.49039149284362793},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.44679901003837585},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.4316745102405548},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.4281182885169983},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.4201643764972687},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.34611210227012634},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.1894223690032959},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.1502009630203247},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.0},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"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/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3472456.3472478","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3472456.3472478","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"50th International Conference on Parallel Processing","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W23806513","https://openalex.org/W993511226","https://openalex.org/W1541729506","https://openalex.org/W1901281515","https://openalex.org/W1987145120","https://openalex.org/W2003117296","https://openalex.org/W2012313745","https://openalex.org/W2014807599","https://openalex.org/W2066166631","https://openalex.org/W2137211879","https://openalex.org/W2159856118","https://openalex.org/W2167868137","https://openalex.org/W2181846018","https://openalex.org/W2402285027","https://openalex.org/W2406344247","https://openalex.org/W2625200202","https://openalex.org/W2761710529","https://openalex.org/W2767308239","https://openalex.org/W2788264912","https://openalex.org/W2789228469","https://openalex.org/W2887282140","https://openalex.org/W2952013349","https://openalex.org/W2985308940","https://openalex.org/W3001467520","https://openalex.org/W3101140947","https://openalex.org/W4206624271","https://openalex.org/W4252602275","https://openalex.org/W4253696084"],"related_works":["https://openalex.org/W2891987081","https://openalex.org/W2983282793","https://openalex.org/W1973046741","https://openalex.org/W1963859303","https://openalex.org/W2364044215","https://openalex.org/W2389600408","https://openalex.org/W240129890","https://openalex.org/W3048701459","https://openalex.org/W2149078538","https://openalex.org/W2370314112"],"abstract_inverted_index":{"Designing":[0],"efficient":[1,68,142],"and":[2,24,61,73,81,123,145,182,205,227,244],"scalable":[3],"sparse":[4],"linear":[5],"algebra":[6],"kernels":[7],"on":[8,89,113,133,176,201,213],"modern":[9],"multi-GPU":[10,97,114,179,249],"based":[11,132],"HPC":[12],"systems":[13,184],"is":[14,237],"a":[15,95,157,202,214],"challenging":[16],"task":[17],"due":[18],"to":[19,58,140,151,197,209,239],"significant":[20],"irregular":[21],"memory":[22,69,105],"references":[23],"workload":[25,74,153],"imbalance":[26],"across":[27],"GPUs.":[28,169],"These":[29],"challenges":[30],"are":[31],"particularly":[32],"compounded":[33],"in":[34,94],"the":[35,109,128,166,177,221,233,242,248],"case":[36],"of":[37,46,111,168,193,247],"Sparse":[38],"Triangular":[39],"Solver":[40],"(SpTRSV),":[41],"which":[42,162],"introduces":[43],"additional":[44],"complexity":[45],"two-dimensional":[47],"computation":[48,52],"dependencies":[49],"among":[50,63],"subsequent":[51],"steps.":[53],"Dependency":[54],"information":[55],"may":[56],"need":[57],"be":[59],"exchanged":[60],"shared":[62],"GPUs,":[64],"thus":[65],"warranting":[66],"for":[67,92],"allocation,":[70],"data":[71],"partitioning,":[72],"distribution":[75],"as":[76,78],"well":[77],"fine-grained":[79,143],"communication":[80,144,245],"synchronization":[82,147],"support.":[83],"In":[84],"this":[85],"work,":[86],"we":[87,126,155],"focus":[88],"designing":[90],"algorithm":[91],"SpTRSV":[93,112,236],"single-node,":[96],"setting.":[98],"We":[99],"demonstrate":[100,185],"that":[101,186,232],"directly":[102],"adopting":[103],"unified":[104],"can":[106,163,189],"adversely":[107],"affect":[108],"performance":[110],"architectures,":[115],"despite":[116],"linking":[117],"via":[118],"fast":[119],"interconnect":[120],"like":[121],"NVLinks":[122],"NVSwitches.":[124],"Alternatively,":[125],"employ":[127],"latest":[129],"NVSHMEM":[130],"technology":[131],"Partitioned":[134],"Global":[135],"Address":[136],"Space":[137],"programming":[138],"model":[139,161],"enable":[141],"drastic":[146],"overhead":[148],"reduction.":[149],"Furthermore,":[150],"handle":[152],"imbalance,":[154],"propose":[156],"malleable":[158],"task-pool":[159],"execution":[160],"further":[164],"enhance":[165],"utilization":[167],"By":[170],"applying":[171],"these":[172],"techniques,":[173],"our":[174,187],"experiments":[175],"NVIDIA":[178],"supernode":[180],"V100-DGX-1":[181],"DGX-2":[183,215],"design":[188],"achieve":[190],"an":[191],"average":[192],"3.53":[194],"\u00d7":[195,207],"(up":[196,208],"9.86":[198],"\u00d7)":[199,211],"speedup":[200,212],"DGX-1":[203],"system":[204,216],"3.66":[206],"9.64":[210],"with":[217],"four":[218],"GPUs":[219],"over":[220],"Unified-Memory":[222],"design.":[223],"The":[224],"comprehensive":[225],"sensitivity":[226],"scalability":[228],"studies":[229],"also":[230],"show":[231],"proposed":[234],"zero-copy":[235],"able":[238],"fully":[240],"utilize":[241],"computing":[243],"resources":[246],"systems.":[250]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":1}],"updated_date":"2026-03-06T13:50:29.536080","created_date":"2025-10-10T00:00:00"}