{"id":"https://openalex.org/W2791833900","doi":"https://doi.org/10.1145/3177964","title":"GPU Performance vs. Thread-Level Parallelism","display_name":"GPU Performance vs. Thread-Level Parallelism","publication_year":2018,"publication_date":"2018-03-22","ids":{"openalex":"https://openalex.org/W2791833900","doi":"https://doi.org/10.1145/3177964","mag":"2791833900"},"language":"en","primary_location":{"id":"doi:10.1145/3177964","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3177964","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":false,"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":null,"license_id":null,"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":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5002035343","display_name":"Zhen Lin","orcid":"https://orcid.org/0000-0002-7131-0159"},"institutions":[{"id":"https://openalex.org/I137902535","display_name":"North Carolina State University","ror":"https://ror.org/04tj63d06","country_code":"US","type":"education","lineage":["https://openalex.org/I137902535"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhen Lin","raw_affiliation_strings":["North Carolina State University, Raleigh, North Carolina"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"North Carolina State University, Raleigh, North Carolina","institution_ids":["https://openalex.org/I137902535"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056016753","display_name":"Michael Mantor","orcid":null},"institutions":[{"id":"https://openalex.org/I4210137977","display_name":"Advanced Micro Devices (United States)","ror":"https://ror.org/04kd6c783","country_code":"US","type":"company","lineage":["https://openalex.org/I4210137977"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Michael Mantor","raw_affiliation_strings":["Advanced Micro Devices, Orlando, Florida"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Advanced Micro Devices, Orlando, Florida","institution_ids":["https://openalex.org/I4210137977"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5072755725","display_name":"Huiyang Zhou","orcid":null},"institutions":[{"id":"https://openalex.org/I137902535","display_name":"North Carolina State University","ror":"https://ror.org/04tj63d06","country_code":"US","type":"education","lineage":["https://openalex.org/I137902535"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Huiyang Zhou","raw_affiliation_strings":["North Carolina State University, Orlando, Florida"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"North Carolina State University, Orlando, Florida","institution_ids":["https://openalex.org/I137902535"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.3835,"has_fulltext":false,"cited_by_count":10,"citation_normalized_percentile":{"value":0.81026759,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":"15","issue":"1","first_page":"1","last_page":"21"},"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":0.9998999834060669,"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":0.9994000196456909,"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":0.8908398151397705},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7623089551925659},{"id":"https://openalex.org/keywords/context-switch","display_name":"Context switch","score":0.6155483722686768},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5522032380104065},{"id":"https://openalex.org/keywords/thread","display_name":"Thread (computing)","score":0.5383113622665405},{"id":"https://openalex.org/keywords/task-parallelism","display_name":"Task parallelism","score":0.46437278389930725},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.35531187057495117},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.33003056049346924},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.19862627983093262},{"id":"https://openalex.org/keywords/parallelism","display_name":"Parallelism (grammar)","score":0.09876129031181335}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8908398151397705},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7623089551925659},{"id":"https://openalex.org/C53833338","wikidata":"https://www.wikidata.org/wiki/Q1061424","display_name":"Context switch","level":2,"score":0.6155483722686768},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5522032380104065},{"id":"https://openalex.org/C138101251","wikidata":"https://www.wikidata.org/wiki/Q213092","display_name":"Thread (computing)","level":2,"score":0.5383113622665405},{"id":"https://openalex.org/C42992933","wikidata":"https://www.wikidata.org/wiki/Q691169","display_name":"Task parallelism","level":3,"score":0.46437278389930725},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.35531187057495117},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.33003056049346924},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.19862627983093262},{"id":"https://openalex.org/C2781172179","wikidata":"https://www.wikidata.org/wiki/Q853109","display_name":"Parallelism (grammar)","level":2,"score":0.09876129031181335}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3177964","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3177964","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":false,"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":null,"license_id":null,"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":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6994652891","display_name":"SHF: Small: Enabling Efficient Context Switching and Effective Latency Hiding in GPUs","funder_award_id":"1618509","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":33,"referenced_works":["https://openalex.org/W1905405560","https://openalex.org/W1968047430","https://openalex.org/W1968775993","https://openalex.org/W1972971542","https://openalex.org/W2047060659","https://openalex.org/W2048441570","https://openalex.org/W2053744175","https://openalex.org/W2079038734","https://openalex.org/W2080592089","https://openalex.org/W2081373884","https://openalex.org/W2081583983","https://openalex.org/W2084110734","https://openalex.org/W2098274770","https://openalex.org/W2110195531","https://openalex.org/W2121523570","https://openalex.org/W2125551452","https://openalex.org/W2128120785","https://openalex.org/W2150851481","https://openalex.org/W2152956697","https://openalex.org/W2163687928","https://openalex.org/W2235785597","https://openalex.org/W2273440736","https://openalex.org/W2323693848","https://openalex.org/W2409690919","https://openalex.org/W2470243357","https://openalex.org/W2510980549","https://openalex.org/W2517837804","https://openalex.org/W2529649238","https://openalex.org/W2554192763","https://openalex.org/W2566334102","https://openalex.org/W2914479353","https://openalex.org/W4243035950","https://openalex.org/W4255847023"],"related_works":["https://openalex.org/W2389214306","https://openalex.org/W2033778626","https://openalex.org/W2965083567","https://openalex.org/W4235240664","https://openalex.org/W2026780467","https://openalex.org/W1838576100","https://openalex.org/W2757182831","https://openalex.org/W2101387113","https://openalex.org/W121182129","https://openalex.org/W4206718021"],"abstract_inverted_index":{"Graphics":[0],"Processing":[1],"Units":[2],"(GPUs)":[3],"leverage":[4],"massive":[5],"thread-level":[6],"parallelism":[7],"(TLP)":[8],"to":[9,52,163,222,240],"achieve":[10],"high":[11],"computation":[12],"throughput":[13,73,117,237],"and":[14,86,92,224,255],"hide":[15],"long":[16,153],"memory":[17,58],"latency.":[18],"However,":[19],"recent":[20],"studies":[21],"have":[22,115],"shown":[23],"that":[24,40,97,112,168,214],"the":[25,32,36,70,87,98,110,156,159,202,208,215,241],"GPU":[26,33,42,77,106],"performance":[27,107,124,216,252],"does":[28],"not":[29,114],"scale":[30],"with":[31,127,235],"occupancy":[34],"or":[35,56,186],"degrees":[37],"of":[38,72,75,84,158,204,228,233],"TLP":[39,132,196,243],"a":[41,65,102,120,139,145,152,169,225,231],"supports,":[43],"especially":[44],"for":[45,105,133,230,257],"memory-intensive":[46],"workloads.":[47],"The":[48,175],"current":[49],"understanding":[50],"points":[51],"L1":[53,90],"D-cache":[54],"contention":[55],"off-chip":[57,80],"bandwidth.":[59],"In":[60],"this":[61,190],"article,":[62],"we":[63,137],"perform":[64],"novel":[66],"scalability":[67],"analysis":[68],"from":[69],"perspective":[71],"utilization":[74,118],"various":[76],"components,":[78],"including":[79],"DRAM,":[81],"multiple":[82],"levels":[83],"caches,":[85],"interconnect":[88,99],"between":[89],"D-caches":[91],"L2":[93],"partitions.":[94],"We":[95],"show":[96],"bandwidth":[100],"is":[101,149,161,178,184],"critical":[103,205],"bound":[104],"scalability.":[108],"For":[109],"applications":[111,135,234],"do":[113],"saturated":[116],"on":[119,253],"particular":[121],"resource,":[122],"their":[123],"scales":[125],"well":[126],"increased":[128],"TLP.":[129],"To":[130],"improve":[131],"such":[134],"efficiently,":[136],"propose":[138],"fast":[140,192],"context":[141,157,193],"switching":[142],"approach.":[143],"When":[144],"warp/thread":[146],"block":[147],"(TB)":[148],"stalled":[150],"by":[151,220],"latency":[154],"operation,":[155],"warp/TB":[160,171,177,183],"spilled":[162],"spare":[164],"on-chip":[165],"resource":[166],"so":[167],"new":[170],"can":[172,197,217],"be":[173,198,218],"launched.":[174],"switched-out":[176],"switched":[179,187],"back":[180],"when":[181],"another":[182],"completed":[185],"out.":[188],"With":[189],"fine-grain":[191],"switching,":[194],"higher":[195,251],"supported":[199],"without":[200],"increasing":[201],"sizes":[203],"resources":[206],"like":[207],"register":[209],"file.":[210],"Our":[211],"experiment":[212],"shows":[213],"improved":[219],"up":[221],"47%":[223],"geometric":[226],"mean":[227],"22%":[229],"set":[232],"unsaturated":[236,258],"utilization.":[238],"Compared":[239],"state-of-the-art":[242],"improvement":[244],"scheme,":[245],"our":[246],"proposed":[247],"scheme":[248],"achieves":[249],"12%":[250],"average":[254],"16%":[256],"benchmarks.":[259]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":2},{"year":2020,"cited_by_count":2},{"year":2019,"cited_by_count":1},{"year":2018,"cited_by_count":1}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}