{"id":"https://openalex.org/W2761710529","doi":"https://doi.org/10.1145/3123939.3124534","title":"Beyond the socket","display_name":"Beyond the socket","publication_year":2017,"publication_date":"2017-10-04","ids":{"openalex":"https://openalex.org/W2761710529","doi":"https://doi.org/10.1145/3123939.3124534","mag":"2761710529"},"language":"en","primary_location":{"id":"doi:10.1145/3123939.3124534","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3123939.3124534","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 50th Annual IEEE/ACM International Symposium on Microarchitecture","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hdl.handle.net/2117/109704","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5076926044","display_name":"Uglje\u0161a Mili\u0107","orcid":null},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]},{"id":"https://openalex.org/I2799803557","display_name":"Barcelona Supercomputing Center","ror":"https://ror.org/05sd8tv96","country_code":"ES","type":"facility","lineage":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":true,"raw_author_name":"Ugljesa Milic","raw_affiliation_strings":["Barcelona Supercomputing Center (BSC) and Universitat Polit\u00e8cnica de Catalunya (UPC)"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC) and Universitat Polit\u00e8cnica de Catalunya (UPC)","institution_ids":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111373927","display_name":"Oreste Villa","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Oreste Villa","raw_affiliation_strings":["NVIDIA"],"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034001619","display_name":"Evgeny Bolotin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Evgeny Bolotin","raw_affiliation_strings":["NVIDIA"],"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087438386","display_name":"Akhil Arunkumar","orcid":null},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Akhil Arunkumar","raw_affiliation_strings":["Arizona State University"],"affiliations":[{"raw_affiliation_string":"Arizona State University","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089789557","display_name":"Eiman Ebrahimi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Eiman Ebrahimi","raw_affiliation_strings":["NVIDIA"],"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078825915","display_name":"Aamer Jaleel","orcid":"https://orcid.org/0000-0002-5709-2992"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aamer Jaleel","raw_affiliation_strings":["NVIDIA"],"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082472991","display_name":"Alex Ram\u00edrez","orcid":"https://orcid.org/0000-0003-3317-4616"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alex Ramirez","raw_affiliation_strings":["Google"],"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5031781240","display_name":"David Nellans","orcid":"https://orcid.org/0000-0001-5203-8367"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"David Nellans","raw_affiliation_strings":["NVIDIA"],"affiliations":[{"raw_affiliation_string":"NVIDIA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5076926044"],"corresponding_institution_ids":["https://openalex.org/I2799803557","https://openalex.org/I9617848"],"apc_list":null,"apc_paid":null,"fwci":7.3971,"has_fulltext":false,"cited_by_count":62,"citation_normalized_percentile":{"value":0.98122769,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"123","last_page":"135"},"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":0.9997000098228455,"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/T10829","display_name":"Interconnection Networks and Systems","score":0.9984999895095825,"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":0.8813961744308472},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.7408236265182495},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.674052357673645},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.45199328660964966},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.4496239125728607},{"id":"https://openalex.org/keywords/interconnection","display_name":"Interconnection","score":0.4383578300476074},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.424377977848053},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.36048421263694763},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3459411859512329},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.17222726345062256}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8813961744308472},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.7408236265182495},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.674052357673645},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.45199328660964966},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.4496239125728607},{"id":"https://openalex.org/C123745756","wikidata":"https://www.wikidata.org/wiki/Q1665949","display_name":"Interconnection","level":2,"score":0.4383578300476074},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.424377977848053},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.36048421263694763},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3459411859512329},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.17222726345062256},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"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/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3123939.3124534","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3123939.3124534","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 50th Annual IEEE/ACM International Symposium on Microarchitecture","raw_type":"proceedings-article"},{"id":"pmh:oai:upcommons.upc.edu:2117/109704","is_oa":true,"landing_page_url":"https://hdl.handle.net/2117/109704","pdf_url":null,"source":{"id":"https://openalex.org/S4210207057","display_name":"QRU Quaderns de Recerca en Urbanisme","issn_l":"2014-9689","issn":["2014-9689","2385-6777"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310322448","host_organization_name":"Q71272178","host_organization_lineage":["https://openalex.org/P4310322448"],"host_organization_lineage_names":["Q71272178"],"type":"journal"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"pmh:oai:upcommons.upc.edu:2117/109704","is_oa":true,"landing_page_url":"https://hdl.handle.net/2117/109704","pdf_url":null,"source":{"id":"https://openalex.org/S4210207057","display_name":"QRU Quaderns de Recerca en Urbanisme","issn_l":"2014-9689","issn":["2014-9689","2385-6777"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310322448","host_organization_name":"Q71272178","host_organization_lineage":["https://openalex.org/P4310322448"],"host_organization_lineage_names":["Q71272178"],"type":"journal"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"},"sustainable_development_goals":[{"score":0.7699999809265137,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G1275801773","display_name":null,"funder_award_id":"BES-2013","funder_id":"https://openalex.org/F4320321837","funder_display_name":"Ministerio de Econom\u00eda y Competitividad"},{"id":"https://openalex.org/G3324247440","display_name":null,"funder_award_id":"TIN2015-65316-P","funder_id":"https://openalex.org/F4320321837","funder_display_name":"Ministerio de Econom\u00eda y Competitividad"},{"id":"https://openalex.org/G5592594182","display_name":null,"funder_award_id":"TIN2015-65316","funder_id":"https://openalex.org/F4320321837","funder_display_name":"Ministerio de Econom\u00eda y Competitividad"},{"id":"https://openalex.org/G8324792492","display_name":null,"funder_award_id":"TIN2012-34557","funder_id":"https://openalex.org/F4320321837","funder_display_name":"Ministerio de Econom\u00eda y Competitividad"}],"funders":[{"id":"https://openalex.org/F4320321837","display_name":"Ministerio de Econom\u00eda y Competitividad","ror":"https://ror.org/034900433"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":36,"referenced_works":["https://openalex.org/W1492601037","https://openalex.org/W1538592187","https://openalex.org/W1555915743","https://openalex.org/W1686810756","https://openalex.org/W1966344015","https://openalex.org/W1968775993","https://openalex.org/W1980419087","https://openalex.org/W1985807728","https://openalex.org/W1988307172","https://openalex.org/W1994999558","https://openalex.org/W1997352364","https://openalex.org/W2014809728","https://openalex.org/W2028173051","https://openalex.org/W2047879375","https://openalex.org/W2067280503","https://openalex.org/W2080592089","https://openalex.org/W2097643185","https://openalex.org/W2125551452","https://openalex.org/W2143773524","https://openalex.org/W2160996172","https://openalex.org/W2167101788","https://openalex.org/W2215665513","https://openalex.org/W2236227338","https://openalex.org/W2281074822","https://openalex.org/W2322798958","https://openalex.org/W2326078278","https://openalex.org/W2327709439","https://openalex.org/W2468407200","https://openalex.org/W2559253174","https://openalex.org/W2566334102","https://openalex.org/W2612094043","https://openalex.org/W2625200202","https://openalex.org/W2770217166","https://openalex.org/W2949245006","https://openalex.org/W4230772827","https://openalex.org/W4236786453"],"related_works":["https://openalex.org/W2378211422","https://openalex.org/W4321353415","https://openalex.org/W2130974462","https://openalex.org/W972276598","https://openalex.org/W2086519370","https://openalex.org/W2028665553","https://openalex.org/W2087343574","https://openalex.org/W2535915176","https://openalex.org/W2105860728","https://openalex.org/W2889526943"],"abstract_inverted_index":{"GPUs":[0,45,179],"achieve":[1,115],"high":[2],"throughput":[3],"and":[4,21,30,99,111,136,155,161,170],"power":[5],"efficiency":[6],"by":[7,152],"employing":[8],"many":[9],"small":[10],"single":[11,150,191],"instruction":[12],"multiple":[13],"thread":[14],"(SIMT)":[15],"cores.":[16],"To":[17],"minimize":[18],"scheduling":[19],"logic":[20],"performance":[22,49,116,188],"variance":[23],"they":[24,57],"utilize":[25],"a":[26,80,149,182,190],"uniform":[27,81],"memory":[28,82,94],"system":[29,83],"leverage":[31],"strong":[32],"data":[33],"parallelism":[34],"exposed":[35],"via":[36],"the":[37,77,108,140],"programming":[38],"model.":[39],"With":[40],"Moore's":[41],"law":[42],"slowing,":[43],"for":[44,185],"to":[46,60,73,106,114,130],"continue":[47],"scaling":[48,186],"(which":[50],"largely":[51],"depends":[52],"on":[53],"SIMT":[54],"core":[55],"count)":[56],"are":[58,66,104],"likely":[59],"embrace":[61],"multi-socket":[62,92,178],"designs":[63,98,173],"where":[64],"transistors":[65],"more":[67],"readily":[68],"available.":[69],"However":[70],"when":[71],"moving":[72],"such":[74],"designs,":[75],"maintaining":[76],"illusion":[78],"of":[79,142,163],"is":[84],"increasingly":[85],"difficult.":[86],"In":[87],"this":[88],"work":[89],"we":[90],"investigate":[91],"non-uniform":[93],"access":[95],"(NUMA)":[96],"GPU":[97,109,128,147,151,187],"show":[100,119],"that":[101,120],"significant":[102],"changes":[103],"needed":[105],"both":[107],"interconnect":[110,135],"cache":[112,137],"architectures":[113],"scalability.":[117],"We":[118],"application":[121,165],"phase":[122],"effects":[123],"can":[124],"be":[125,181],"exploited":[126],"allowing":[127],"sockets":[129,172],"dynamically":[131],"optimize":[132],"their":[133],"individual":[134],"policies,":[138],"minimizing":[139],"impact":[141],"NUMA":[143],"effects.":[144],"Our":[145],"NUMA-aware":[146,177],"outperforms":[148],"1.5\u00d7,":[153],"2.3\u00d7,":[154],"3.2\u00d7":[156],"while":[157],"achieving":[158],"89%,":[159],"84%,":[160],"76%":[162],"theoretical":[164],"scalability":[166],"in":[167],"2,":[168],"4,":[169],"8":[171],"respectively.":[174],"Implementable":[175],"today,":[176],"may":[180],"promising":[183],"candidate":[184],"beyond":[189],"socket.":[192]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":7},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":12},{"year":2020,"cited_by_count":12},{"year":2019,"cited_by_count":17},{"year":2018,"cited_by_count":3}],"updated_date":"2026-04-10T15:06:20.359241","created_date":"2017-10-20T00:00:00"}
