{"id":"https://openalex.org/W3102175148","doi":"https://doi.org/10.1145/3424669","title":"SMAUG","display_name":"SMAUG","publication_year":2020,"publication_date":"2020-11-10","ids":{"openalex":"https://openalex.org/W3102175148","doi":"https://doi.org/10.1145/3424669","mag":"3102175148"},"language":"en","primary_location":{"id":"doi:10.1145/3424669","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3424669","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3424669","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://dl.acm.org/doi/pdf/10.1145/3424669","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5053699612","display_name":"Sam Likun Xi","orcid":null},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Sam (Likun) Xi","raw_affiliation_strings":["Harvard University, Cambridge, MA"],"affiliations":[{"raw_affiliation_string":"Harvard University, Cambridge, MA","institution_ids":["https://openalex.org/I2801851002"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081046846","display_name":"Yuan Yao","orcid":"https://orcid.org/0000-0001-9448-5595"},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yuan Yao","raw_affiliation_strings":["Harvard University, Cambridge, MA"],"affiliations":[{"raw_affiliation_string":"Harvard University, Cambridge, MA","institution_ids":["https://openalex.org/I2801851002"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072298152","display_name":"Kshitij Bhardwaj","orcid":"https://orcid.org/0000-0001-7076-9251"},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kshitij Bhardwaj","raw_affiliation_strings":["Harvard University, Cambridge, MA"],"affiliations":[{"raw_affiliation_string":"Harvard University, Cambridge, MA","institution_ids":["https://openalex.org/I2801851002"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103266146","display_name":"Paul N. Whatmough","orcid":null},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Paul Whatmough","raw_affiliation_strings":["Harvard University and Arm ML Research, Cambridge, MA"],"affiliations":[{"raw_affiliation_string":"Harvard University and Arm ML Research, Cambridge, MA","institution_ids":["https://openalex.org/I136199984"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043327132","display_name":"Gu-Yeon Wei","orcid":"https://orcid.org/0000-0001-5730-9904"},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gu-Yeon Wei","raw_affiliation_strings":["Harvard University, Cambridge, MA"],"affiliations":[{"raw_affiliation_string":"Harvard University, Cambridge, MA","institution_ids":["https://openalex.org/I2801851002"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5026496503","display_name":"David Brooks","orcid":"https://orcid.org/0000-0002-0662-7889"},"institutions":[{"id":"https://openalex.org/I2801851002","display_name":"Harvard University Press","ror":"https://ror.org/006v7bf86","country_code":"US","type":"other","lineage":["https://openalex.org/I136199984","https://openalex.org/I2801851002"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David Brooks","raw_affiliation_strings":["Harvard University, Cambridge, MA"],"affiliations":[{"raw_affiliation_string":"Harvard University, Cambridge, MA","institution_ids":["https://openalex.org/I2801851002"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5053699612"],"corresponding_institution_ids":["https://openalex.org/I2801851002"],"apc_list":null,"apc_paid":null,"fwci":3.9255,"has_fulltext":true,"cited_by_count":60,"citation_normalized_percentile":{"value":0.94949194,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":100},"biblio":{"volume":"17","issue":"4","first_page":"1","last_page":"26"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9991999864578247,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9990000128746033,"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":0.8496346473693848},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.65128493309021},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6153745651245117},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5916794538497925},{"id":"https://openalex.org/keywords/microarchitecture","display_name":"Microarchitecture","score":0.58536297082901},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.5290443301200867},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.5269152522087097},{"id":"https://openalex.org/keywords/network-topology","display_name":"Network topology","score":0.47899672389030457},{"id":"https://openalex.org/keywords/arm-architecture","display_name":"ARM architecture","score":0.4554774761199951},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.4367803931236267},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.43426191806793213},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.4204982817173004},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.38294923305511475},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.36150234937667847},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3487478196620941},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.3455185890197754},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.1957467496395111},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.1402752697467804},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.10186171531677246}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8496346473693848},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.65128493309021},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6153745651245117},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5916794538497925},{"id":"https://openalex.org/C107598950","wikidata":"https://www.wikidata.org/wiki/Q259864","display_name":"Microarchitecture","level":2,"score":0.58536297082901},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.5290443301200867},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.5269152522087097},{"id":"https://openalex.org/C199845137","wikidata":"https://www.wikidata.org/wiki/Q145490","display_name":"Network topology","level":2,"score":0.47899672389030457},{"id":"https://openalex.org/C26771161","wikidata":"https://www.wikidata.org/wiki/Q16980","display_name":"ARM architecture","level":2,"score":0.4554774761199951},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.4367803931236267},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.43426191806793213},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.4204982817173004},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.38294923305511475},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.36150234937667847},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3487478196620941},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3455185890197754},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.1957467496395111},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.1402752697467804},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.10186171531677246},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3424669","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3424669","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3424669","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:10.1145/3424669","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3424669","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3424669","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":[{"id":"https://metadata.un.org/sdg/7","score":0.8299999833106995,"display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G1047302985","display_name":null,"funder_award_id":"1718160","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G3417205766","display_name":null,"funder_award_id":"CNS-1718160","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4493967758","display_name":null,"funder_award_id":"1533737","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8724954132","display_name":null,"funder_award_id":"1718160","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320306087","display_name":"Semiconductor Research Corporation","ror":"https://ror.org/047z4n946"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3102175148.pdf","grobid_xml":"https://content.openalex.org/works/W3102175148.grobid-xml"},"referenced_works_count":53,"referenced_works":["https://openalex.org/W2005487033","https://openalex.org/W2055312318","https://openalex.org/W2076089313","https://openalex.org/W2094756095","https://openalex.org/W2095076490","https://openalex.org/W2112678088","https://openalex.org/W2112796928","https://openalex.org/W2155893237","https://openalex.org/W2158190559","https://openalex.org/W2194775991","https://openalex.org/W2285660444","https://openalex.org/W2406293284","https://openalex.org/W2508602506","https://openalex.org/W2513554817","https://openalex.org/W2515287984","https://openalex.org/W2518281301","https://openalex.org/W2605516844","https://openalex.org/W2606722458","https://openalex.org/W2612076670","https://openalex.org/W2727238169","https://openalex.org/W2751366252","https://openalex.org/W2762155252","https://openalex.org/W2789554134","https://openalex.org/W2790925711","https://openalex.org/W2794598778","https://openalex.org/W2808970127","https://openalex.org/W2899915146","https://openalex.org/W2932154853","https://openalex.org/W2940862705","https://openalex.org/W2944824859","https://openalex.org/W2946225357","https://openalex.org/W2963387679","https://openalex.org/W2963709005","https://openalex.org/W2979334884","https://openalex.org/W2980186997","https://openalex.org/W2980200167","https://openalex.org/W2997929983","https://openalex.org/W2998228662","https://openalex.org/W2998732502","https://openalex.org/W3012178976","https://openalex.org/W3024621361","https://openalex.org/W3097528158","https://openalex.org/W3102385154","https://openalex.org/W3104393472","https://openalex.org/W3104463312","https://openalex.org/W4230706651","https://openalex.org/W4247198796","https://openalex.org/W4249553981","https://openalex.org/W4249932213","https://openalex.org/W4251430913","https://openalex.org/W4288083528","https://openalex.org/W4302408070","https://openalex.org/W6713184358"],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2532502681","https://openalex.org/W3127172286","https://openalex.org/W4255097844","https://openalex.org/W2567660513","https://openalex.org/W3211218493","https://openalex.org/W2463263055","https://openalex.org/W3015480182","https://openalex.org/W2774142458","https://openalex.org/W2384439204"],"abstract_inverted_index":{"In":[0],"recent":[1],"years,":[2],"there":[3,86],"has":[4,20,68],"been":[5,69],"tremendous":[6],"advances":[7],"in":[8,59,106],"hardware":[9,99],"acceleration":[10],"of":[11,17,121,132,155,185],"deep":[12,61,123,202],"neural":[13],"networks.":[14],"However,":[15],"most":[16],"the":[18,44,52,60,112,151,186],"research":[19,107],"focused":[21],"on":[22,32,55],"optimizing":[23],"accelerator":[24,45,100,144,187],"microarchitecture":[25],"for":[26,39,119,134,172,199],"higher":[27],"performance":[28,76,168],"and":[29,58,146,153,169],"energy":[30,170],"efficiency":[31,171],"a":[33,129,178,200],"per-layer":[34],"basis.":[35],"We":[36],"find":[37],"that":[38,92,116,161],"overall":[40,167],"single-batch":[41],"inference":[42],"latency,":[43],"may":[46],"only":[47],"make":[48],"up":[49,173],"25\u201340%,":[50],"with":[51,96],"rest":[53],"spent":[54],"data":[56],"movement":[57],"learning":[62,124,203],"software":[63],"framework.":[64],"Thus":[65],"far,":[66],"it":[67],"very":[70],"difficult":[71],"to":[72,142,174],"study":[73],"end-to-end":[74,94,122],"DNN":[75,90,114,136],"during":[77],"early":[78],"stage":[79],"design":[80],"(before":[81],"RTL":[82],"is":[83,117],"available),":[84],"because":[85],"are":[87],"no":[88],"existing":[89],"frameworks":[91],"support":[93],"simulation":[95,120],"easy":[97,143],"custom":[98],"integration.":[101,148],"To":[102,149],"address":[103],"this":[104],"gap":[105],"infrastructure,":[108],"we":[109,157,164],"present":[110,158],"SMAUG,":[111,156],"first":[113],"framework":[115],"purpose-built":[118],"applications.":[125],"SMAUG":[126,194],"offers":[127],"researchers":[128],"wide":[130],"range":[131],"capabilities":[133],"evaluating":[135],"workloads,":[137],"from":[138],"diverse":[139],"network":[140],"topologies":[141],"modeling":[145],"SoC":[147,198],"demonstrate":[150],"power":[152],"value":[154],"case":[159],"studies":[160],"show":[162,192],"how":[163,193],"can":[165,195],"optimize":[166],"1.8\u00d7\u20135\u00d7":[175],"speedup":[176],"over":[177],"baseline":[179],"system,":[180],"without":[181],"changing":[182],"any":[183],"part":[184],"microarchitecture,":[188],"as":[189,191],"well":[190],"tune":[196],"an":[197],"camera-powered":[201],"pipeline.":[204]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":10},{"year":2023,"cited_by_count":9},{"year":2022,"cited_by_count":12},{"year":2021,"cited_by_count":16},{"year":2020,"cited_by_count":3}],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2020-11-23T00:00:00"}