{"id":"https://openalex.org/W4381327256","doi":"https://doi.org/10.1145/3577193.3593713","title":"Multi-GPU Communication Schemes for Iterative Solvers: When CPUs are Not in Charge","display_name":"Multi-GPU Communication Schemes for Iterative Solvers: When CPUs are Not in Charge","publication_year":2023,"publication_date":"2023-06-20","ids":{"openalex":"https://openalex.org/W4381327256","doi":"https://doi.org/10.1145/3577193.3593713"},"language":"en","primary_location":{"id":"doi:10.1145/3577193.3593713","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3577193.3593713","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3577193.3593713","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 37th 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/10.1145/3577193.3593713","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5071088229","display_name":"Ismayil Ismayilov","orcid":"https://orcid.org/0000-0003-1831-5556"},"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":"Ismayil Ismayilov","raw_affiliation_strings":["Ko\u00e7 University, Istanbul, Turkey"],"raw_orcid":"https://orcid.org/0000-0003-1831-5556","affiliations":[{"raw_affiliation_string":"Ko\u00e7 University, Istanbul, Turkey","institution_ids":["https://openalex.org/I1351752"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092212133","display_name":"Javid Baydamirli","orcid":"https://orcid.org/0000-0001-7235-6418"},"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":"Javid Baydamirli","raw_affiliation_strings":["Ko\u00e7 University, Istanbul, Turkey"],"raw_orcid":"https://orcid.org/0000-0001-7235-6418","affiliations":[{"raw_affiliation_string":"Ko\u00e7 University, Istanbul, Turkey","institution_ids":["https://openalex.org/I1351752"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092212134","display_name":"Do\u011fan Sa\u011fbili","orcid":"https://orcid.org/0000-0002-9603-2466"},"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":"Do\u011fan Sa\u011fbili","raw_affiliation_strings":["Ko\u00e7 University, Istanbul, Turkey"],"raw_orcid":"https://orcid.org/0000-0002-9603-2466","affiliations":[{"raw_affiliation_string":"Ko\u00e7 University, Istanbul, Turkey","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, Kobe, Japan"],"raw_orcid":"https://orcid.org/0000-0002-7165-2095","affiliations":[{"raw_affiliation_string":"RIKEN Center for Computational Science, Kobe, 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, Turkey"],"raw_orcid":"https://orcid.org/0000-0002-2351-0770","affiliations":[{"raw_affiliation_string":"Ko\u00e7 University, Istanbul, Turkey","institution_ids":["https://openalex.org/I1351752"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5071088229"],"corresponding_institution_ids":["https://openalex.org/I1351752"],"apc_list":null,"apc_paid":null,"fwci":4.7706,"has_fulltext":true,"cited_by_count":16,"citation_normalized_percentile":{"value":0.95960982,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"192","last_page":"202"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9998999834060669,"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":0.9998999834060669,"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/T10715","display_name":"Distributed and Parallel Computing 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":0.9980000257492065,"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":0.8673447370529175},{"id":"https://openalex.org/keywords/stencil","display_name":"Stencil","score":0.7985855937004089},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6408964395523071},{"id":"https://openalex.org/keywords/porting","display_name":"Porting","score":0.6363725066184998},{"id":"https://openalex.org/keywords/central-processing-unit","display_name":"Central processing unit","score":0.5622081756591797},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5547189712524414},{"id":"https://openalex.org/keywords/compiler","display_name":"Compiler","score":0.512309193611145},{"id":"https://openalex.org/keywords/software-portability","display_name":"Software portability","score":0.49246057868003845},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.45724013447761536},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.43463706970214844},{"id":"https://openalex.org/keywords/execution-model","display_name":"Execution model","score":0.4276992082595825},{"id":"https://openalex.org/keywords/thread","display_name":"Thread (computing)","score":0.4175490736961365},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.26990288496017456},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.15260273218154907},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.1323314607143402}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8673447370529175},{"id":"https://openalex.org/C76752949","wikidata":"https://www.wikidata.org/wiki/Q7607499","display_name":"Stencil","level":2,"score":0.7985855937004089},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6408964395523071},{"id":"https://openalex.org/C106251023","wikidata":"https://www.wikidata.org/wiki/Q851989","display_name":"Porting","level":3,"score":0.6363725066184998},{"id":"https://openalex.org/C49154492","wikidata":"https://www.wikidata.org/wiki/Q5300","display_name":"Central processing unit","level":2,"score":0.5622081756591797},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5547189712524414},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.512309193611145},{"id":"https://openalex.org/C63000827","wikidata":"https://www.wikidata.org/wiki/Q3080428","display_name":"Software portability","level":2,"score":0.49246057868003845},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.45724013447761536},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.43463706970214844},{"id":"https://openalex.org/C2776834041","wikidata":"https://www.wikidata.org/wiki/Q25346349","display_name":"Execution model","level":2,"score":0.4276992082595825},{"id":"https://openalex.org/C138101251","wikidata":"https://www.wikidata.org/wiki/Q213092","display_name":"Thread (computing)","level":2,"score":0.4175490736961365},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.26990288496017456},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.15260273218154907},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.1323314607143402},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3577193.3593713","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3577193.3593713","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3577193.3593713","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 37th International Conference on Supercomputing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3577193.3593713","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3577193.3593713","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3577193.3593713","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 37th 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"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4381327256.pdf","grobid_xml":"https://content.openalex.org/works/W4381327256.grobid-xml"},"referenced_works_count":36,"referenced_works":["https://openalex.org/W1943052934","https://openalex.org/W1967821347","https://openalex.org/W1988756252","https://openalex.org/W1991709203","https://openalex.org/W2023415862","https://openalex.org/W2035080386","https://openalex.org/W2054522323","https://openalex.org/W2056295840","https://openalex.org/W2096714979","https://openalex.org/W2121504627","https://openalex.org/W2134427337","https://openalex.org/W2171399035","https://openalex.org/W2292731977","https://openalex.org/W2402285027","https://openalex.org/W2482071863","https://openalex.org/W2521121845","https://openalex.org/W2612387305","https://openalex.org/W2617808919","https://openalex.org/W2623223928","https://openalex.org/W2730016893","https://openalex.org/W2753515784","https://openalex.org/W2767210925","https://openalex.org/W2767308239","https://openalex.org/W2778540992","https://openalex.org/W2785822536","https://openalex.org/W2898048683","https://openalex.org/W2998812396","https://openalex.org/W3043110088","https://openalex.org/W3046421657","https://openalex.org/W3163459356","https://openalex.org/W4205468814","https://openalex.org/W4229701516","https://openalex.org/W4242172941","https://openalex.org/W4248478717","https://openalex.org/W4285100150","https://openalex.org/W4316252357"],"related_works":["https://openalex.org/W3105129168","https://openalex.org/W2356602486","https://openalex.org/W2804920739","https://openalex.org/W4316371992","https://openalex.org/W2621501241","https://openalex.org/W2208757713","https://openalex.org/W3177128669","https://openalex.org/W2991848348","https://openalex.org/W3184765594","https://openalex.org/W2357149307"],"abstract_inverted_index":{"This":[0],"paper":[1],"proposes":[2],"a":[3,25,47,150],"fully":[4,105],"autonomous":[5,106],"execution":[6,36,84],"model":[7,119,139],"for":[8,49,153],"multi-GPU":[9,27,107],"applications":[10,74],"that":[11,53,75],"completely":[12],"excludes":[13],"the":[14,17,20,29,33,81,134,137],"involvement":[15],"of":[16,35],"CPU":[18],"beyond":[19],"initial":[21],"kernel":[22],"launch.":[23],"In":[24],"typical":[26],"application,":[28],"host":[30],"serves":[31],"as":[32,46,91],"orchestrator":[34],"by":[37,146],"directly":[38],"launching":[39],"kernels,":[40,93],"issuing":[41],"communication":[42,77,101,113,144],"calls,":[43],"and":[44,63,99,109,129,148],"acting":[45],"synchronizer":[48],"devices.":[50],"We":[51,115],"argue":[52],"this":[54],"orchestration,":[55],"or":[56],"control":[57],"flow":[58],"path,":[59],"causes":[60],"undue":[61],"overhead":[62],"can":[64,140],"be":[65],"delegated":[66],"entirely":[67],"to":[68,70,103,133],"devices":[69],"improve":[71,141],"performance":[72],"in":[73],"require":[76],"among":[78],"peers.":[79],"For":[80],"proposed":[82,118],"CPU-free":[83,138],"model,":[85],"we":[86],"leverage":[87],"existing":[88],"techniques":[89],"such":[90],"persistent":[92],"thread":[94],"block":[95],"specialization,":[96],"device-side":[97],"barriers,":[98],"device-initiated":[100],"routines":[102],"write":[104],"code":[108,162],"achieve":[110],"significantly":[111],"reduced":[112],"overheads.":[114],"demonstrate":[116],"our":[117],"on":[120,155],"two":[121],"broadly":[122],"used":[123],"iterative":[124],"solvers,":[125],"2D/3D":[126],"Jacobi":[127],"stencil":[128,143],"Conjugate":[130],"Gradient(CG).":[131],"Compared":[132],"CPU-controlled":[135],"baselines,":[136],"3D":[142],"latency":[145],"58.8%":[147],"provide":[149],"1.63x":[151],"speedup":[152],"CG":[154],"8":[156],"NVIDIA":[157],"A100":[158],"GPUs.":[159],"The":[160],"project":[161],"is":[163],"available":[164],"at":[165],"https://github.com/ParCoreLab/CPU-Free-model.":[166]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":7},{"year":2023,"cited_by_count":1}],"updated_date":"2026-05-04T08:30:34.212998","created_date":"2025-10-10T00:00:00"}