{"id":"https://openalex.org/W4413412046","doi":"https://doi.org/10.1145/3721145.3730426","title":"A Device-Side Execution Model for Multi-GPU Task Graphs","display_name":"A Device-Side Execution Model for Multi-GPU Task Graphs","publication_year":2025,"publication_date":"2025-06-08","ids":{"openalex":"https://openalex.org/W4413412046","doi":"https://doi.org/10.1145/3721145.3730426"},"language":"en","primary_location":{"id":"doi:10.1145/3721145.3730426","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3721145.3730426","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3721145.3730426","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 39th ACM International Conference on Supercomputing","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/10.1145/3721145.3730426","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5049538914","display_name":"Ilyas Turimbetov","orcid":"https://orcid.org/0000-0002-0384-6330"},"institutions":[{"id":"https://openalex.org/I1351752","display_name":"Ko\u00e7 University","ror":"https://ror.org/00jzwgz36","country_code":"TR","type":"education","lineage":["https://openalex.org/I1351752"]}],"countries":["TR"],"is_corresponding":true,"raw_author_name":"Ilyas Turimbetov","raw_affiliation_strings":["Ko\u00e7 University, Istanbul, Turkiye"],"affiliations":[{"raw_affiliation_string":"Ko\u00e7 University, Istanbul, Turkiye","institution_ids":["https://openalex.org/I1351752"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002208999","display_name":"Mohamed Wahib","orcid":"https://orcid.org/0000-0002-7165-2095"},"institutions":[{"id":"https://openalex.org/I4210129730","display_name":"RIKEN Center for Computational Science","ror":"https://ror.org/03r519674","country_code":"JP","type":"facility","lineage":["https://openalex.org/I4210110652","https://openalex.org/I4210129730"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Mohamed Wahib","raw_affiliation_strings":["RIKEN Center for Computational Science, Tokyo, Japan"],"affiliations":[{"raw_affiliation_string":"RIKEN Center for Computational Science, Tokyo, Japan","institution_ids":["https://openalex.org/I4210129730"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5040713791","display_name":"Didem Unat","orcid":"https://orcid.org/0000-0002-2351-0770"},"institutions":[{"id":"https://openalex.org/I1351752","display_name":"Ko\u00e7 University","ror":"https://ror.org/00jzwgz36","country_code":"TR","type":"education","lineage":["https://openalex.org/I1351752"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"Didem Unat","raw_affiliation_strings":["Ko\u00e7 University, Istanbul, Turkiye"],"affiliations":[{"raw_affiliation_string":"Ko\u00e7 University, Istanbul, Turkiye","institution_ids":["https://openalex.org/I1351752"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5049538914"],"corresponding_institution_ids":["https://openalex.org/I1351752"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.22237086,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"384","last_page":"396"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9991000294685364,"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":0.9991000294685364,"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/T12292","display_name":"Graph Theory and Algorithms","score":0.9991000294685364,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.8422589302062988},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6696515679359436},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6168292164802551}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8422589302062988},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6696515679359436},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6168292164802551},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3721145.3730426","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3721145.3730426","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3721145.3730426","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 39th ACM International Conference on Supercomputing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3721145.3730426","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3721145.3730426","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3721145.3730426","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 39th ACM International Conference on Supercomputing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G4114410456","display_name":"Pioneering a New Path in Parallel Programming Beyond Moore\u2019s Law","funder_award_id":"949587","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G4956428346","display_name":null,"funder_award_id":"Horizon 2020 research and innovatio","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G7057791530","display_name":null,"funder_award_id":"949587","funder_id":"https://openalex.org/F4320334678","funder_display_name":"European Research Council"},{"id":"https://openalex.org/G7842005466","display_name":null,"funder_award_id":"Horizon 2020","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320334678","display_name":"European Research Council","ror":"https://ror.org/0472cxd90"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4413412046.pdf","grobid_xml":"https://content.openalex.org/works/W4413412046.grobid-xml"},"referenced_works_count":30,"referenced_works":["https://openalex.org/W1991709203","https://openalex.org/W1996511977","https://openalex.org/W2087440962","https://openalex.org/W2134427337","https://openalex.org/W2148153323","https://openalex.org/W2171226522","https://openalex.org/W2402285027","https://openalex.org/W2581090187","https://openalex.org/W2626009569","https://openalex.org/W2730016893","https://openalex.org/W2767210925","https://openalex.org/W2768826296","https://openalex.org/W2805009828","https://openalex.org/W2891759099","https://openalex.org/W2913704315","https://openalex.org/W2954427512","https://openalex.org/W2962686325","https://openalex.org/W2983126917","https://openalex.org/W2998104365","https://openalex.org/W3038173700","https://openalex.org/W3174311009","https://openalex.org/W3187688612","https://openalex.org/W4242172941","https://openalex.org/W4248722156","https://openalex.org/W4318541614","https://openalex.org/W4367684987","https://openalex.org/W4381327256","https://openalex.org/W4393931335","https://openalex.org/W4405755154","https://openalex.org/W4406163815"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Executing":[0],"task":[1,32,70,89,130,137],"graphs":[2,71,90,105],"on":[3,72,109,143],"multi-GPU":[4,53,81],"systems":[5],"presents":[6],"challenges":[7],"typically":[8],"managed":[9],"by":[10],"CPU-side":[11],"runtimes,":[12],"which":[13],"handle":[14],"memory":[15],"management,":[16],"track":[17],"dependencies,":[18],"and":[19,30,61,139,163,201],"balance":[20],"load.However,":[21],"the":[22,93,120,129,144,147,180,190],"interplay":[23],"of":[24,87,122,136],"runtime":[25,124],"components,":[26],"CPUdriven":[27],"kernel":[28,117],"initialization,":[29],"dynamic":[31],"graph":[33],"construction":[34],"creates":[35],"significant":[36],"overhead.For":[37],"static":[38,88],"graphs,":[39,158],"recent":[40],"advancements":[41],"have":[42],"enabled":[43],"GPU-side":[44,65],"execution,":[45],"demonstrating":[46],"substantial":[47],"performance":[48],"gains":[49],"in":[50,58],"single-GPU":[51],"scenarios.However,":[52],"execution":[54,82,86],"still":[55],"lags":[56],"behind":[57],"both":[59],"usability":[60],"performance.In":[62],"particular,":[63],"no":[64],"solution":[66,101],"exists":[67],"for":[68,102,149,199,203],"executing":[69,103],"multiple":[73,107,110],"nodes.In":[74],"this":[75],"work,":[76],"we":[77],"introduce":[78],"Mustard,":[79],"a":[80,99,167,185],"model":[83],"that":[84],"shifts":[85],"entirely":[91],"to":[92,115,189],"devices,":[94],"drastically":[95],"reducing":[96],"overhead.Mustard":[97],"offers":[98],"clean":[100],"CUDA":[104],"across":[106],"GPUs":[108],"nodes":[111],"without":[112],"requiring":[113],"modifications":[114],"GPU":[116],"code":[118],"or":[119,126],"adoption":[121],"new":[123],"mechanisms":[125],"APIs.By":[127],"transforming":[128],"graph,":[131],"Mustard":[132,173,193],"enables":[133],"precise":[134],"tracking":[135],"dependencies":[138],"load":[140],"balancing":[141],"directly":[142],"GPU,":[145],"eliminating":[146],"need":[148],"host":[150],"CPU":[151],"involvement.We":[152],"evaluate":[153],"our":[154],"approach":[155],"using":[156],"generated":[157],"as":[159,161],"well":[160],"LU":[162,200],"Cholesky":[164],"decomposition":[165],"graphs.In":[166],"multi-node":[168],"scenario":[169],"with":[170],"64":[171],"GPUs,":[172],"achieves":[174],"an":[175,195],"average":[176,196],"5.83":[177],"speedup":[178,198],"over":[179],"linear":[181],"algebra":[182],"library":[183],"SLATE.On":[184],"single":[186],"node,":[187],"compared":[188],"best-performing":[191],"baseline,":[192],"delivers":[194],"1.66":[197],"1.29":[202],"Cholesky.":[204]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-10-10T00:00:00"}