{"id":"https://openalex.org/W2625954420","doi":"https://doi.org/10.1145/3061639.3062207","title":"Automated Systolic Array Architecture Synthesis for High Throughput CNN Inference on FPGAs","display_name":"Automated Systolic Array Architecture Synthesis for High Throughput CNN Inference on FPGAs","publication_year":2017,"publication_date":"2017-06-13","ids":{"openalex":"https://openalex.org/W2625954420","doi":"https://doi.org/10.1145/3061639.3062207","mag":"2625954420"},"language":"en","primary_location":{"id":"doi:10.1145/3061639.3062207","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3061639.3062207","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th Annual Design Automation Conference 2017","raw_type":"proceedings-article"},"type":"conference-paper","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/A5102904614","display_name":"Xuechao Wei","orcid":"https://orcid.org/0000-0002-0996-2260"},"institutions":[{"id":"https://openalex.org/I4210143490","display_name":"Computing Center","ror":"https://ror.org/0557kgc34","country_code":"RU","type":"facility","lineage":["https://openalex.org/I1313323035","https://openalex.org/I4210143490","https://openalex.org/I4210148470"]}],"countries":["RU"],"is_corresponding":false,"raw_author_name":"Xuechao Wei","raw_affiliation_strings":["Center for Energy-efficient Computing and Applications, School of EECS, Peking University, China and Falcon Computing Solutions, Inc, Los Angeles, CA, USA and Falcon Computing, Solutions Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Energy-efficient Computing and Applications, School of EECS, Peking University, China and Falcon Computing Solutions, Inc, Los Angeles, CA, USA and Falcon Computing, Solutions Inc","institution_ids":["https://openalex.org/I4210143490"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008633310","display_name":"Cody Hao Yu","orcid":"https://orcid.org/0000-0002-9298-6254"},"institutions":[{"id":"https://openalex.org/I4210113436","display_name":"SC Solutions (United States)","ror":"https://ror.org/01x5py295","country_code":"US","type":"company","lineage":["https://openalex.org/I4210113436"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Cody Hao Yu","raw_affiliation_strings":["Computer Science Department, University of California, Los Angeles, CA, USA and Falcon Computing Solutions, Inc, Los Angeles, CA, USA and Falcon Computing, Solutions Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Computer Science Department, University of California, Los Angeles, CA, USA and Falcon Computing Solutions, Inc, Los Angeles, CA, USA and Falcon Computing, Solutions Inc","institution_ids":["https://openalex.org/I4210113436"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101949155","display_name":"Peng Zhang","orcid":"https://orcid.org/0000-0001-7721-2675"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng Zhang","raw_affiliation_strings":["Falcon Computing Solutions, Inc, Los Angeles, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Falcon Computing Solutions, Inc, Los Angeles, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091638904","display_name":"Youxiang Chen","orcid":"https://orcid.org/0000-0002-5744-1312"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Youxiang Chen","raw_affiliation_strings":["Falcon Computing Solutions, Inc, Los Angeles, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Falcon Computing Solutions, Inc, Los Angeles, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101597230","display_name":"Yuxin Wang","orcid":"https://orcid.org/0000-0002-1481-0608"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuxin Wang","raw_affiliation_strings":["Falcon Computing Solutions, Inc, Los Angeles, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Falcon Computing Solutions, Inc, Los Angeles, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091049278","display_name":"Hu Han","orcid":"https://orcid.org/0000-0001-6010-1792"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han Hu","raw_affiliation_strings":["Falcon Computing Solutions, Inc, Los Angeles, CA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Falcon Computing Solutions, Inc, Los Angeles, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100604860","display_name":"Yun Liang","orcid":"https://orcid.org/0000-0002-9076-7998"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yun Liang","raw_affiliation_strings":["Center for Energy-efficient Computing and Applications, School of EECS, Peking University, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Energy-efficient Computing and Applications, School of EECS, Peking University, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016776689","display_name":"Jason Cong","orcid":"https://orcid.org/0000-0003-2887-6963"},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]},{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Jason Cong","raw_affiliation_strings":["Center for Energy-efficient Computing and Applications, School of EECS, Peking University, China and Computer Science Department, University of California, Los Angeles, CA, USA and Falcon Computing Solutions, Inc, Los Angeles, CA, USA and Falcon Computing Solutions Inc. and Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Energy-efficient Computing and Applications, School of EECS, Peking University, China and Computer Science Department, University of California, Los Angeles, CA, USA and Falcon Computing Solutions, Inc, Los Angeles, CA, USA and Falcon Computing Solutions Inc. and Peking University","institution_ids":["https://openalex.org/I20231570","https://openalex.org/I161318765"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":418,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9986000061035156,"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":0.9986000061035156,"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/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9962000250816345,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.9896000027656555,"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/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.8313068151473999},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7855361700057983},{"id":"https://openalex.org/keywords/flops","display_name":"FLOPS","score":0.7216653823852539},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6937688589096069},{"id":"https://openalex.org/keywords/systolic-array","display_name":"Systolic array","score":0.6788508892059326},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.5872839689254761},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.5438888072967529},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.5367053747177124},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.5120165944099426},{"id":"https://openalex.org/keywords/adder","display_name":"Adder","score":0.5064424276351929},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.495571494102478},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4619479775428772},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4491375684738159},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4222950041294098},{"id":"https://openalex.org/keywords/clock-rate","display_name":"Clock rate","score":0.41622650623321533},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.343711256980896},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2767770290374756},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.20069193840026855},{"id":"https://openalex.org/keywords/very-large-scale-integration","display_name":"Very-large-scale integration","score":0.15521171689033508},{"id":"https://openalex.org/keywords/chip","display_name":"Chip","score":0.15349242091178894}],"concepts":[{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.8313068151473999},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7855361700057983},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.7216653823852539},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6937688589096069},{"id":"https://openalex.org/C150741067","wikidata":"https://www.wikidata.org/wiki/Q2377218","display_name":"Systolic array","level":3,"score":0.6788508892059326},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.5872839689254761},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.5438888072967529},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.5367053747177124},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.5120165944099426},{"id":"https://openalex.org/C164620267","wikidata":"https://www.wikidata.org/wiki/Q376953","display_name":"Adder","level":3,"score":0.5064424276351929},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.495571494102478},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4619479775428772},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4491375684738159},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4222950041294098},{"id":"https://openalex.org/C178693496","wikidata":"https://www.wikidata.org/wiki/Q911691","display_name":"Clock rate","level":3,"score":0.41622650623321533},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.343711256980896},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2767770290374756},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.20069193840026855},{"id":"https://openalex.org/C14580979","wikidata":"https://www.wikidata.org/wiki/Q876049","display_name":"Very-large-scale integration","level":2,"score":0.15521171689033508},{"id":"https://openalex.org/C165005293","wikidata":"https://www.wikidata.org/wiki/Q1074500","display_name":"Chip","level":2,"score":0.15349242091178894},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.0},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3061639.3062207","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3061639.3062207","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th Annual Design Automation Conference 2017","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.7099999785423279,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W1487564550","https://openalex.org/W1500590019","https://openalex.org/W1558370006","https://openalex.org/W1686810756","https://openalex.org/W1990315422","https://openalex.org/W2009832130","https://openalex.org/W2053968820","https://openalex.org/W2076066374","https://openalex.org/W2094756095","https://openalex.org/W2096645269","https://openalex.org/W2097117768","https://openalex.org/W2117696986","https://openalex.org/W2139092702","https://openalex.org/W2155893237","https://openalex.org/W2163605009","https://openalex.org/W2276486856","https://openalex.org/W2294282016","https://openalex.org/W2513568085","https://openalex.org/W2520083297","https://openalex.org/W2574797063","https://openalex.org/W2584311934","https://openalex.org/W2584616277","https://openalex.org/W2585774018","https://openalex.org/W2618530766","https://openalex.org/W2729080111"],"related_works":["https://openalex.org/W4390550886","https://openalex.org/W3217463396","https://openalex.org/W2790557758","https://openalex.org/W3037505396","https://openalex.org/W2560894929","https://openalex.org/W2156524298","https://openalex.org/W2106942255","https://openalex.org/W1492102884","https://openalex.org/W1894233513","https://openalex.org/W2137485365"],"abstract_inverted_index":{"Convolutional":[0],"neural":[1],"networks":[2],"(CNNs)":[3],"have":[4,36],"been":[5],"widely":[6],"applied":[7],"in":[8],"many":[9],"deep":[10],"learning":[11],"applications.":[12],"In":[13,48],"recent":[14],"years,":[15],"the":[16,41,45,118],"FPGA":[17,56],"implementation":[18,103],"for":[19,77,120,129,137],"CNNs":[20],"has":[21],"attracted":[22],"much":[23],"attention":[24],"because":[25],"of":[26,44],"its":[27],"high":[28,65,69],"performance":[29,78],"and":[30,79,82,134],"energy":[31],"efficiency.":[32],"However,":[33],"existing":[34],"implementations":[35],"difficulty":[37],"to":[38,100,116,126],"fully":[39],"leverage":[40],"computation":[42],"power":[43],"latest":[46],"FPGAs.":[47],"this":[49],"paper":[50],"we":[51],"implement":[52],"CNN":[53,102,122],"on":[54],"an":[55,74,84],"using":[57,104],"a":[58,97,101],"systolic":[59,105],"array":[60],"architecture,":[61],"which":[62],"can":[63],"achieve":[64],"clock":[66],"frequency":[67],"under":[68],"resource":[70,80],"utilization.":[71],"We":[72],"provide":[73],"analytical":[75],"model":[76],"utilization":[81],"develop":[83],"automatic":[85],"design":[86],"space":[87],"exploration":[88],"framework,":[89],"as":[90,92],"well":[91],"source-to-source":[93],"code":[94],"transformation":[95],"from":[96],"C":[98],"program":[99],"array.":[106],"The":[107],"experimental":[108],"results":[109],"show":[110],"that":[111],"our":[112],"framework":[113],"is":[114],"able":[115],"generate":[117],"accelerator":[119],"real-life":[121],"models,":[123],"achieving":[124],"up":[125],"461":[127],"GFlops":[128],"floating":[130],"point":[131],"data":[132],"type":[133],"1.2":[135],"Tops":[136],"8-16":[138],"bit":[139],"fixed":[140],"point.":[141]},"counts_by_year":[{"year":2026,"cited_by_count":8},{"year":2025,"cited_by_count":28},{"year":2024,"cited_by_count":36},{"year":2023,"cited_by_count":31},{"year":2022,"cited_by_count":45},{"year":2021,"cited_by_count":76},{"year":2020,"cited_by_count":83},{"year":2019,"cited_by_count":54},{"year":2018,"cited_by_count":51},{"year":2017,"cited_by_count":6}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}