{"id":"https://openalex.org/W2625200202","doi":"https://doi.org/10.1145/3079856.3080231","title":"MCM-GPU","display_name":"MCM-GPU","publication_year":2017,"publication_date":"2017-06-15","ids":{"openalex":"https://openalex.org/W2625200202","doi":"https://doi.org/10.1145/3079856.3080231","mag":"2625200202"},"language":"en","primary_location":{"id":"doi:10.1145/3079856.3080231","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3079856.3080231","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 44th Annual International Symposium on Computer Architecture","raw_type":"proceedings-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/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":true,"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/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/A5041232825","display_name":"Benjamin Cho","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":"Benjamin Cho","raw_affiliation_strings":["University of Texas at Austin"],"affiliations":[{"raw_affiliation_string":"University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076926044","display_name":"Uglje\u0161a Mili\u0107","orcid":null},"institutions":[{"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"]},{"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"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Ugljesa Milic","raw_affiliation_strings":["Barcelona Supercomputing Center / Universitat Politecnica de Catalunya"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center / Universitat Politecnica de Catalunya","institution_ids":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]}]},{"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/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/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/A5028220093","display_name":"Carole-Jean Wu","orcid":"https://orcid.org/0000-0002-9032-7239"},"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":"Carole-Jean Wu","raw_affiliation_strings":["Arizona State University"],"affiliations":[{"raw_affiliation_string":"Arizona State University","institution_ids":["https://openalex.org/I55732556"]}]},{"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":9,"corresponding_author_ids":["https://openalex.org/A5087438386"],"corresponding_institution_ids":["https://openalex.org/I55732556"],"apc_list":null,"apc_paid":null,"fwci":18.9265,"has_fulltext":false,"cited_by_count":158,"citation_normalized_percentile":{"value":0.99702346,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"320","last_page":"332"},"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":0.9994999766349792,"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/T12326","display_name":"Network Packet Processing and Optimization","score":0.9980000257492065,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8357832431793213},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.7328850030899048},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.6720185279846191},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6545008420944214},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.5997312664985657},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.5368196964263916},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.5354136824607849},{"id":"https://openalex.org/keywords/locality","display_name":"Locality","score":0.5037469267845154},{"id":"https://openalex.org/keywords/general-purpose-computing-on-graphics-processing-units","display_name":"General-purpose computing on graphics processing units","score":0.4837629795074463},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.4722425639629364},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.4443742632865906},{"id":"https://openalex.org/keywords/transistor","display_name":"Transistor","score":0.42974111437797546},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.3530275523662567},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.2711167633533478},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.21514326333999634},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.08019888401031494},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.07768213748931885}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8357832431793213},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.7328850030899048},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.6720185279846191},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6545008420944214},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.5997312664985657},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.5368196964263916},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.5354136824607849},{"id":"https://openalex.org/C2779808786","wikidata":"https://www.wikidata.org/wiki/Q6664603","display_name":"Locality","level":2,"score":0.5037469267845154},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.4837629795074463},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4722425639629364},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.4443742632865906},{"id":"https://openalex.org/C172385210","wikidata":"https://www.wikidata.org/wiki/Q5339","display_name":"Transistor","level":3,"score":0.42974111437797546},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3530275523662567},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2711167633533478},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.21514326333999634},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.08019888401031494},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.07768213748931885},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","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/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3079856.3080231","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3079856.3080231","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 44th Annual International Symposium on Computer Architecture","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.46000000834465027}],"awards":[],"funders":[{"id":"https://openalex.org/F4320309480","display_name":"Nvidia","ror":"https://ror.org/03jdj4y14"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W1966344015","https://openalex.org/W1985539519","https://openalex.org/W1985807728","https://openalex.org/W1988307172","https://openalex.org/W2024122052","https://openalex.org/W2029764709","https://openalex.org/W2035794753","https://openalex.org/W2067280503","https://openalex.org/W2079038734","https://openalex.org/W2080592089","https://openalex.org/W2092823305","https://openalex.org/W2093852742","https://openalex.org/W2103838950","https://openalex.org/W2121444153","https://openalex.org/W2144558071","https://openalex.org/W2148153323","https://openalex.org/W2153801723","https://openalex.org/W2157802978","https://openalex.org/W2161190431","https://openalex.org/W2164645802","https://openalex.org/W2166984288","https://openalex.org/W2167101788","https://openalex.org/W2169665207","https://openalex.org/W2172654076","https://openalex.org/W2234584938","https://openalex.org/W2344289226","https://openalex.org/W2396363708","https://openalex.org/W2464584271","https://openalex.org/W2492136742","https://openalex.org/W2511683765","https://openalex.org/W4255962840"],"related_works":["https://openalex.org/W1963859303","https://openalex.org/W2364044215","https://openalex.org/W2389600408","https://openalex.org/W240129890","https://openalex.org/W3048701459","https://openalex.org/W2149078538","https://openalex.org/W2080146221","https://openalex.org/W2370314112","https://openalex.org/W1912958759","https://openalex.org/W2792081825"],"abstract_inverted_index":{"Historically,":[0],"improvements":[1],"in":[2,52,59],"GPU-based":[3],"high":[4,102],"performance":[5,33,79],"computing":[6],"have":[7],"been":[8],"tightly":[9],"coupled":[10],"to":[11,50,71,161],"transistor":[12],"scaling.":[13],"As":[14],"Moore's":[15,82],"law":[16],"slows":[17],"down,":[18],"and":[19,96,104,114,137,154,181,215],"the":[20,32,43,112,116,139,148,162,168,176,209],"number":[21,212],"of":[22,35,67,118,185,213],"transistors":[23],"per":[24],"die":[25],"no":[26],"longer":[27],"grows":[28],"at":[29],"historical":[30],"rates,":[31],"curve":[34],"single":[36],"monolithic":[37,179,190],"GPUs":[38,48,75,88],"will":[39],"ultimately":[40],"plateau.":[41],"However,":[42],"need":[44],"for":[45],"higher":[46],"performing":[47],"continues":[49],"exist":[51],"many":[53],"domains.":[54],"To":[55],"address":[56],"this":[57,60],"need,":[58],"paper":[61],"we":[62,85,193],"demonstrate":[63],"that":[64,131,147,195],"package-level":[65],"integration":[66],"multiple":[68],"GPU":[69,93,122],"modules":[70],"build":[72],"larger":[73],"logical":[74],"can":[76],"enable":[77],"continuous":[78],"scaling":[80],"beyond":[81],"law.":[83],"Specifically,":[84],"propose":[86,127],"partitioning":[87],"into":[89],"easily":[90],"manufacturable":[91],"basic":[92,120,163],"Modules":[94],"(GPMs),":[95],"integrating":[97],"them":[98],"on":[99,141],"package":[100],"using":[101],"bandwidth":[103,157],"power":[105],"efficient":[106],"signaling":[107],"technologies.":[108],"We":[109,125],"lay":[110],"out":[111],"details":[113],"evaluate":[115],"feasibility":[117],"a":[119,186],"Multi-Chip-Module":[121],"(MCM-GPU)":[123],"design.":[124],"then":[126],"three":[128],"architectural":[129],"optimizations":[130],"significantly":[132],"improve":[133],"GPM":[134],"data":[135],"locality":[136],"minimize":[138],"sensitivity":[140],"inter-GPM":[142,156],"bandwidth.":[143,217],"Our":[144],"evaluation":[145],"shows":[146],"optimized":[149,169,197],"MCM-GPU":[150,164,170,198],"achieves":[151],"22.8%":[152],"speedup":[153],"5x":[155],"reduction":[158],"when":[159],"compared":[160],"architecture.":[165],"Most":[166],"importantly,":[167],"design":[171],"is":[172,199],"45.5%":[173],"faster":[174,201],"than":[175,202],"largest":[177],"implementable":[178],"GPU,":[180],"performs":[182],"within":[183],"10%":[184],"hypothetical":[187],"(and":[188],"unbuildable)":[189],"GPU.":[191],"Lastly":[192],"show":[194],"our":[196],"26.8%":[200],"an":[203],"equally":[204],"equipped":[205],"Multi-GPU":[206],"system":[207],"with":[208],"same":[210],"total":[211],"SMs":[214],"DRAM":[216]},"counts_by_year":[{"year":2025,"cited_by_count":14},{"year":2024,"cited_by_count":14},{"year":2023,"cited_by_count":14},{"year":2022,"cited_by_count":11},{"year":2021,"cited_by_count":21},{"year":2020,"cited_by_count":26},{"year":2019,"cited_by_count":37},{"year":2018,"cited_by_count":17},{"year":2017,"cited_by_count":4}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2017-06-23T00:00:00"}