{"id":"https://openalex.org/W4411670522","doi":"https://doi.org/10.1145/3746231","title":"Performance Implications of Pipelining the Data Transfer in CPU-GPU Heterogeneous Systems","display_name":"Performance Implications of Pipelining the Data Transfer in CPU-GPU Heterogeneous Systems","publication_year":2025,"publication_date":"2025-06-26","ids":{"openalex":"https://openalex.org/W4411670522","doi":"https://doi.org/10.1145/3746231"},"language":"en","primary_location":{"id":"doi:10.1145/3746231","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746231","pdf_url":null,"source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1145/3746231","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101477549","display_name":"Ruihao Li","orcid":"https://orcid.org/0000-0002-7092-2401"},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ruihao Li","raw_affiliation_strings":["Electrical and Computer Engineering, The University of Texas at Austin","Electrical and Computer Engineering, The University of Texas at Austin, Austin, United States"],"raw_orcid":"https://orcid.org/0000-0002-7092-2401","affiliations":[{"raw_affiliation_string":"Electrical and Computer Engineering, The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]},{"raw_affiliation_string":"Electrical and Computer Engineering, The University of Texas at Austin, Austin, United States","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019583227","display_name":"Bagus Hanindhito","orcid":"https://orcid.org/0000-0002-8485-581X"},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bagus Hanindhito","raw_affiliation_strings":["The University of Texas at Austin","The University of Texas at Austin, Austin, United States"],"raw_orcid":"https://orcid.org/0000-0002-8485-581X","affiliations":[{"raw_affiliation_string":"The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]},{"raw_affiliation_string":"The University of Texas at Austin, Austin, United States","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112976084","display_name":"Sanjana Yadav","orcid":null},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sanjana Yadav","raw_affiliation_strings":["The University of Texas at Austin","The University of Texas at Austin, Austin, United States"],"raw_orcid":"https://orcid.org/0009-0007-7880-9386","affiliations":[{"raw_affiliation_string":"The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]},{"raw_affiliation_string":"The University of Texas at Austin, Austin, United States","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078434282","display_name":"Qinzhe Wu","orcid":"https://orcid.org/0000-0002-7988-1431"},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qinzhe Wu","raw_affiliation_strings":["The University of Texas at Austin","The University of Texas at Austin, Austin, United States"],"raw_orcid":"https://orcid.org/0000-0002-7988-1431","affiliations":[{"raw_affiliation_string":"The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]},{"raw_affiliation_string":"The University of Texas at Austin, Austin, United States","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004353960","display_name":"Krishna M. Kavi","orcid":"https://orcid.org/0000-0002-0876-166X"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Krishna Kavi","raw_affiliation_strings":["University of North Texas","University of North Texas, Denton, United States"],"raw_orcid":"https://orcid.org/0000-0002-0876-166X","affiliations":[{"raw_affiliation_string":"University of North Texas","institution_ids":["https://openalex.org/I123534392"]},{"raw_affiliation_string":"University of North Texas, Denton, United States","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078470359","display_name":"Gayatri Mehta","orcid":"https://orcid.org/0000-0001-7754-1874"},"institutions":[{"id":"https://openalex.org/I123534392","display_name":"University of North Texas","ror":"https://ror.org/00v97ad02","country_code":"US","type":"education","lineage":["https://openalex.org/I123534392"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gayatri Mehta","raw_affiliation_strings":["University of North Texas","University of North Texas, Denton, United States"],"raw_orcid":"https://orcid.org/0000-0001-7754-1874","affiliations":[{"raw_affiliation_string":"University of North Texas","institution_ids":["https://openalex.org/I123534392"]},{"raw_affiliation_string":"University of North Texas, Denton, United States","institution_ids":["https://openalex.org/I123534392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031510918","display_name":"Neeraja J. Yadwadkar","orcid":"https://orcid.org/0009-0007-7556-3069"},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Neeraja J. Yadwadkar","raw_affiliation_strings":["The University of Texas at Austin","The University of Texas at Austin, Austin, United States"],"raw_orcid":"https://orcid.org/0009-0007-7556-3069","affiliations":[{"raw_affiliation_string":"The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]},{"raw_affiliation_string":"The University of Texas at Austin, Austin, United States","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5068885069","display_name":"Lizy K. John","orcid":"https://orcid.org/0000-0002-8747-5214"},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lizy K. John","raw_affiliation_strings":["The University of Texas at Austin","The University of Texas at Austin, Austin, United States"],"raw_orcid":"https://orcid.org/0000-0002-8747-5214","affiliations":[{"raw_affiliation_string":"The University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]},{"raw_affiliation_string":"The University of Texas at Austin, Austin, United States","institution_ids":["https://openalex.org/I86519309"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.754,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.89390445,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":"22","issue":"3","first_page":"1","last_page":"26"},"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9973999857902527,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9969000220298767,"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":0.8463760614395142},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7296339869499207},{"id":"https://openalex.org/keywords/supercomputer","display_name":"Supercomputer","score":0.4357251822948456},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.42684459686279297}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8463760614395142},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7296339869499207},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.4357251822948456},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.42684459686279297}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746231","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746231","pdf_url":null,"source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1145/3746231","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746231","pdf_url":null,"source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W2080592089","https://openalex.org/W2100415730","https://openalex.org/W2624632852","https://openalex.org/W2794945502","https://openalex.org/W2912055016","https://openalex.org/W2930869794","https://openalex.org/W3011293047","https://openalex.org/W3028661980","https://openalex.org/W3093933627","https://openalex.org/W3125884253","https://openalex.org/W3130554079","https://openalex.org/W3173778203","https://openalex.org/W3182263566","https://openalex.org/W3205144645","https://openalex.org/W3205803342","https://openalex.org/W3209568355","https://openalex.org/W4221032334","https://openalex.org/W4237687284","https://openalex.org/W4242764150","https://openalex.org/W4301361180","https://openalex.org/W4320067882","https://openalex.org/W4361197980","https://openalex.org/W4387321091","https://openalex.org/W4394907705","https://openalex.org/W4394998532"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2384867379","https://openalex.org/W4400094300","https://openalex.org/W2329539859","https://openalex.org/W1599154403","https://openalex.org/W1582436825","https://openalex.org/W1996803181","https://openalex.org/W2979588510"],"abstract_inverted_index":{"Driven":[0],"by":[1,361],"the":[2,17,54,82,112,121,142,156,175,307,315,334,343,357],"increasing":[3],"demands":[4],"of":[5,151,174,178,194,205,219,246,306,318,336],"machine":[6],"learning,":[7],"heterogeneous":[8],"systems":[9],"combining":[10],"CPUs":[11,34],"and":[12,30,35,47,63,159,183,210,224,260,266,338,368],"GPUs":[13,38,109],"have":[14,39],"emerged":[15],"as":[16,340,342],"dominant":[18],"architecture":[19],"for":[20],"parallel":[21],"computing":[22],"in":[23,81,107],"recent":[24],"years.":[25],"To":[26],"optimize":[27],"memory":[28,43,49,61,98,127,130,258,345,363],"management":[29],"data":[31,71,76,122,144,162,239,323,358,366],"transfer":[32,123,145,163,240,324,359,367],"between":[33],"GPUs,":[36],"Nvidia":[37,108],"introduced":[40,105],"unified":[41],"virtual":[42],"(":[44,50,100],"UVM":[45,57,158,179,223,279],")":[46,52,103],"pinned":[48],"PM":[51,73,160,181,230,281],"over":[53,277],"last":[55],"decade.":[56],"can":[58,131],"avoid":[59],"explicit":[60],"copies":[62],"potentially":[64],"overlap":[65],"GPU":[66,125,135,308,371],"kernel":[67,309],"computations":[68],"with":[69,77,134,365],"CPU-GPU":[70,113,143,161],"transfer.":[72],"ensures":[74],"that":[75],"high":[78,261],"locality":[79],"remains":[80],"main":[83],"memory,":[84],"preventing":[85],"it":[86],"from":[87,124,191],"being":[88],"paged":[89],"out.":[90],"In":[91,167,283],"addition":[92],"to":[93,110,128,141,235,313,331],"these":[94,195,200],"two":[95],"techniques,":[96],"asynchronous":[97],"copy":[99],"Async":[101,118,153,184,247,269,297],"Memcpy":[102,119,154,185,248,270,298],"was":[104],"recently":[106],"improve":[111],"pipeline":[114,360],"further.":[115],"By":[116],"utilizing":[117,222],",":[120,180,182,186,288],"global":[126],"shared":[129,257],"be":[132],"overlapped":[133],"computations,":[136],"adding":[137],"an":[138,215,303],"additional":[139],"stage":[140],"pipeline.":[146],"A":[147],"thorough":[148],"performance":[149,172,217,244,275,293,311,319],"analysis":[150],"how":[152,333],"affects":[155],"current":[157],"scheme":[164],"is":[165],"desired.":[166],"this":[168],"article,":[169],"we":[170,289,301],"provide":[171],"implications":[173],"combined":[176],"effect":[177],"exploring":[187],"which":[188,192],"applications":[189],"benefit":[190],"combination":[193],"features.":[196],"We":[197,213,326,349],"implement":[198],"all":[199],"features":[201],"on":[202,231],"a":[203,225,273,291],"suite":[204],"25":[206],"workloads,":[207],"including":[208],"microbenchmarks":[209],"realworld":[211,232],"applications.":[212],"observe":[214],"average":[216],"gain":[218,227],"24%":[220],"when":[221,228,295],"34%":[226],"employing":[229],"applications,":[233],"compared":[234],"not":[236],"applying":[237],"any":[238],"optimization":[241],"techniques.":[242],"The":[243],"benefits":[245],"vary":[249],"across":[250,370],"different":[251],"workloads.":[252],"For":[253],"workloads":[254,285],"featuring":[255],"extensive":[256],"usage":[259],"compute":[262],"density":[263],"(e.g.,":[264],"kmeans":[265],"lud":[267],"),":[268],"delivers":[271],"around":[272],"20%":[274,292],"improvement":[276],"using":[278,296,310],"or":[280],"alone.":[282],"other":[284],"like":[286],"knn":[287],"note":[290],"degradation":[294],".":[299],"Furthermore,":[300],"conduct":[302],"in-depth":[304],"investigation":[305],"counters":[312],"uncover":[314],"root":[316],"causes":[317],"differences":[320],"among":[321],"various":[322],"models.":[325],"also":[327],"perform":[328],"sensitivity":[329],"analyses":[330],"examine":[332],"number":[335],"blocks":[337],"threads,":[339],"well":[341],"L1-cache/shared":[344],"partitioning,":[346],"impact":[347],"performance.":[348],"explore":[350],"future":[351],"research":[352],"directions":[353],"aimed":[354],"at":[355],"enhancing":[356],"overlapping":[362],"allocation":[364],"computation":[369],"kernels.":[372]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}