{"id":"https://openalex.org/W2964108906","doi":"https://doi.org/10.1109/vts.2018.8368656","title":"Analyzing and mitigating the impact of permanent faults on a systolic array based neural network accelerator","display_name":"Analyzing and mitigating the impact of permanent faults on a systolic array based neural network accelerator","publication_year":2018,"publication_date":"2018-04-01","ids":{"openalex":"https://openalex.org/W2964108906","doi":"https://doi.org/10.1109/vts.2018.8368656","mag":"2964108906"},"language":"en","primary_location":{"id":"doi:10.1109/vts.2018.8368656","is_oa":false,"landing_page_url":"https://doi.org/10.1109/vts.2018.8368656","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 IEEE 36th VLSI Test Symposium (VTS)","raw_type":"proceedings-article"},"type":"preprint","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/A5077249620","display_name":"Jeff Zhang","orcid":"https://orcid.org/0000-0001-7411-8923"},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jeff Jun Zhang","raw_affiliation_strings":["Department of Electrical and Computer Engineering, New York University, Brooklyn, New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, New York University, Brooklyn, New York","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025675252","display_name":"Tianyu Gu","orcid":"https://orcid.org/0000-0001-5134-3896"},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tianyu Gu","raw_affiliation_strings":["Department of Electrical and Computer Engineering, New York University, Brooklyn, New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, New York University, Brooklyn, New York","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066320524","display_name":"Kanad Basu","orcid":"https://orcid.org/0000-0002-6431-7512"},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kanad Basu","raw_affiliation_strings":["Department of Electrical and Computer Engineering, New York University, Brooklyn, New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, New York University, Brooklyn, New York","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5010950688","display_name":"Siddharth Garg","orcid":"https://orcid.org/0000-0002-6158-9512"},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Siddharth Garg","raw_affiliation_strings":["Department of Electrical and Computer Engineering, New York University, Brooklyn, New York"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, New York University, Brooklyn, New York","institution_ids":["https://openalex.org/I57206974"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I57206974"],"apc_list":null,"apc_paid":null,"fwci":17.8139,"has_fulltext":false,"cited_by_count":171,"citation_normalized_percentile":{"value":0.99688474,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9997000098228455,"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":0.9997000098228455,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9977999925613403,"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/T10829","display_name":"Interconnection Networks and Systems","score":0.996399998664856,"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/systolic-array","display_name":"Systolic array","score":0.7613805532455444},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.717766284942627},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5901916027069092},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5456394553184509},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.527273952960968},{"id":"https://openalex.org/keywords/fault","display_name":"Fault (geology)","score":0.4189072251319885},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4050734043121338},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.36013317108154297},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.35811299085617065},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2013138234615326},{"id":"https://openalex.org/keywords/very-large-scale-integration","display_name":"Very-large-scale integration","score":0.08671194314956665}],"concepts":[{"id":"https://openalex.org/C150741067","wikidata":"https://www.wikidata.org/wiki/Q2377218","display_name":"Systolic array","level":3,"score":0.7613805532455444},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.717766284942627},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5901916027069092},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5456394553184509},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.527273952960968},{"id":"https://openalex.org/C175551986","wikidata":"https://www.wikidata.org/wiki/Q47089","display_name":"Fault (geology)","level":2,"score":0.4189072251319885},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4050734043121338},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.36013317108154297},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.35811299085617065},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2013138234615326},{"id":"https://openalex.org/C14580979","wikidata":"https://www.wikidata.org/wiki/Q876049","display_name":"Very-large-scale integration","level":2,"score":0.08671194314956665},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C165205528","wikidata":"https://www.wikidata.org/wiki/Q83371","display_name":"Seismology","level":1,"score":0.0},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/vts.2018.8368656","is_oa":false,"landing_page_url":"https://doi.org/10.1109/vts.2018.8368656","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 IEEE 36th VLSI Test Symposium (VTS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":44,"referenced_works":["https://openalex.org/W1937359183","https://openalex.org/W1965729545","https://openalex.org/W1971368502","https://openalex.org/W2016053056","https://openalex.org/W2017369466","https://openalex.org/W2017521824","https://openalex.org/W2067523571","https://openalex.org/W2083547484","https://openalex.org/W2088079057","https://openalex.org/W2100956485","https://openalex.org/W2119144962","https://openalex.org/W2121881146","https://openalex.org/W2128941141","https://openalex.org/W2130189691","https://openalex.org/W2130494464","https://openalex.org/W2130942839","https://openalex.org/W2134797427","https://openalex.org/W2139777941","https://openalex.org/W2145942961","https://openalex.org/W2147435261","https://openalex.org/W2149637086","https://openalex.org/W2155604655","https://openalex.org/W2161033118","https://openalex.org/W2163605009","https://openalex.org/W2186094539","https://openalex.org/W2276892413","https://openalex.org/W2289252105","https://openalex.org/W2346555444","https://openalex.org/W2560615381","https://openalex.org/W2606722458","https://openalex.org/W2625028270","https://openalex.org/W2625840880","https://openalex.org/W2657126969","https://openalex.org/W2707890836","https://openalex.org/W2962965870","https://openalex.org/W2964299589","https://openalex.org/W3139765438","https://openalex.org/W4244330903","https://openalex.org/W4247470470","https://openalex.org/W6678071927","https://openalex.org/W6679436768","https://openalex.org/W6679909955","https://openalex.org/W6684191040","https://openalex.org/W6704402508"],"related_works":["https://openalex.org/W4240320454","https://openalex.org/W2070314832","https://openalex.org/W2395557210","https://openalex.org/W2010232134","https://openalex.org/W2384485649","https://openalex.org/W3184376546","https://openalex.org/W2373300491","https://openalex.org/W2395294869","https://openalex.org/W2378744544","https://openalex.org/W2565094479"],"abstract_inverted_index":{"Due":[0],"to":[1,104,111,152],"their":[2],"growing":[3],"popularity":[4],"and":[5,96,123],"computational":[6],"cost,":[7],"deep":[8],"neural":[9],"networks":[10],"(DNNs)":[11],"are":[12],"being":[13],"targeted":[14],"for":[15,21,56],"hardware":[16],"acceleration.":[17],"A":[18],"popular":[19],"architecture":[20],"DNN":[22,54],"acceleration,":[23],"adopted":[24],"by":[25],"the":[26,47,68,102,162,166],"Google":[27],"Tensor":[28],"Processing":[29],"Unit":[30],"(TPU),":[31],"utilizes":[32],"a":[33,72,132],"systolic":[34,51],"array":[35,52],"based":[36,53],"matrix":[37],"multiplication":[38],"unit":[39],"at":[40,78,106],"its":[41],"core.":[42],"This":[43],"paper":[44],"deals":[45],"with":[46,113],"design":[48],"of":[49,71,109,165],"fault-tolerant,":[50],"accelerators":[55],"high":[57],"defect":[58],"rate":[59],"technologies.":[60],"To":[61],"this":[62,149],"end,":[63],"we":[64,144],"empirically":[65],"show":[66],"that":[67,100,147],"classification":[69,117],"accuracy":[70,118],"baseline":[73],"TPU":[74,103,137],"drops":[75],"significantly":[76],"even":[77],"extremely":[79],"low":[80,84,120],"fault":[81,107],"rates":[82,108],"(as":[83,119],"as":[85,121],"0.006%).":[86],"We":[87],"then":[88,159],"propose":[89,145],"two":[90],"novel":[91],"strategies,":[92],"fault-aware":[93,97],"pruning":[94],"(FAP)":[95],"pruning+retraining":[98],"(FAP+T),":[99],"enable":[101],"operate":[105],"up":[110],"50%,":[112],"negligible":[114],"drop":[115],"in":[116],"0.1%)":[122],"no":[124],"run-time":[125],"performance":[126],"overhead.":[127],"The":[128,156],"FAP+T":[129],"does":[130],"introduce":[131],"one-time":[133,150],"retraining":[134],"penalty":[135,151,157],"per":[136],"chip":[138],"before":[139],"it":[140],"is":[141,158],"deployed,":[142],"but":[143],"optimizations":[146],"reduce":[148],"under":[153],"12":[154],"minutes.":[155],"amortized":[160],"over":[161],"entire":[163],"lifetime":[164],"TPU's":[167],"operation.":[168]},"counts_by_year":[{"year":2026,"cited_by_count":7},{"year":2025,"cited_by_count":15},{"year":2024,"cited_by_count":22},{"year":2023,"cited_by_count":30},{"year":2022,"cited_by_count":22},{"year":2021,"cited_by_count":35},{"year":2020,"cited_by_count":19},{"year":2019,"cited_by_count":13},{"year":2018,"cited_by_count":8}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}