{"id":"https://openalex.org/W4386765150","doi":"https://doi.org/10.1109/dac56929.2023.10247793","title":"Accelerating DNN Inference with Heterogeneous Multi-DPU Engines","display_name":"Accelerating DNN Inference with Heterogeneous Multi-DPU Engines","publication_year":2023,"publication_date":"2023-07-09","ids":{"openalex":"https://openalex.org/W4386765150","doi":"https://doi.org/10.1109/dac56929.2023.10247793"},"language":"en","primary_location":{"id":"doi:10.1109/dac56929.2023.10247793","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac56929.2023.10247793","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 60th ACM/IEEE Design Automation Conference (DAC)","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/A5109684348","display_name":"Zelin Du","orcid":null},"institutions":[{"id":"https://openalex.org/I80143920","display_name":"Shandong University of Science and Technology","ror":"https://ror.org/04gtjhw98","country_code":"CN","type":"education","lineage":["https://openalex.org/I80143920"]},{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Zelin Du","raw_affiliation_strings":["Shandong University,School of Cyber Science and Technology,Qingdao,China","Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong","School of Cyber Science and Technology, Shandong University, Qingdao, China"],"affiliations":[{"raw_affiliation_string":"Shandong University,School of Cyber Science and Technology,Qingdao,China","institution_ids":["https://openalex.org/I80143920"]},{"raw_affiliation_string":"Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]},{"raw_affiliation_string":"School of Cyber Science and Technology, Shandong University, Qingdao, China","institution_ids":["https://openalex.org/I80143920"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000814233","display_name":"Wei Zhang","orcid":"https://orcid.org/0000-0002-0195-2666"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]},{"id":"https://openalex.org/I80143920","display_name":"Shandong University of Science and Technology","ror":"https://ror.org/04gtjhw98","country_code":"CN","type":"education","lineage":["https://openalex.org/I80143920"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Zhang","raw_affiliation_strings":["Shandong University,School of Cyber Science and Technology,Qingdao,China","School of Cyber Science and Technology, Shandong University, Qingdao, China","Quan Cheng Laboratory, Jinan, China"],"affiliations":[{"raw_affiliation_string":"Shandong University,School of Cyber Science and Technology,Qingdao,China","institution_ids":["https://openalex.org/I80143920"]},{"raw_affiliation_string":"School of Cyber Science and Technology, Shandong University, Qingdao, China","institution_ids":["https://openalex.org/I80143920"]},{"raw_affiliation_string":"Quan Cheng Laboratory, Jinan, China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025167300","display_name":"Zimeng Zhou","orcid":"https://orcid.org/0000-0002-6683-3991"},"institutions":[{"id":"https://openalex.org/I80143920","display_name":"Shandong University of Science and Technology","ror":"https://ror.org/04gtjhw98","country_code":"CN","type":"education","lineage":["https://openalex.org/I80143920"]},{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zimeng Zhou","raw_affiliation_strings":["Shandong University,School of Cyber Science and Technology,Qingdao,China","School of Cyber Science and Technology, Shandong University, Qingdao, China","Quan Cheng Laboratory, Jinan, China"],"affiliations":[{"raw_affiliation_string":"Shandong University,School of Cyber Science and Technology,Qingdao,China","institution_ids":["https://openalex.org/I80143920"]},{"raw_affiliation_string":"School of Cyber Science and Technology, Shandong University, Qingdao, China","institution_ids":["https://openalex.org/I80143920"]},{"raw_affiliation_string":"Quan Cheng Laboratory, Jinan, China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101639532","display_name":"Zili Shao","orcid":"https://orcid.org/0000-0002-2173-2847"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zili Shao","raw_affiliation_strings":["The Chinese University of Hong Kong,Department of Computer Science and Engineering,Hong Kong","Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong"],"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Department of Computer Science and Engineering,Hong Kong","institution_ids":["https://openalex.org/I177725633"]},{"raw_affiliation_string":"Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5063104755","display_name":"Lei Ju","orcid":"https://orcid.org/0000-0001-6186-5399"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]},{"id":"https://openalex.org/I80143920","display_name":"Shandong University of Science and Technology","ror":"https://ror.org/04gtjhw98","country_code":"CN","type":"education","lineage":["https://openalex.org/I80143920"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lei Ju","raw_affiliation_strings":["Shandong University,School of Cyber Science and Technology,Qingdao,China","Quan Cheng Laboratory, Jinan, China","School of Cyber Science and Technology, Shandong University, Qingdao, China"],"affiliations":[{"raw_affiliation_string":"Shandong University,School of Cyber Science and Technology,Qingdao,China","institution_ids":["https://openalex.org/I80143920"]},{"raw_affiliation_string":"Quan Cheng Laboratory, Jinan, China","institution_ids":["https://openalex.org/I4210136793"]},{"raw_affiliation_string":"School of Cyber Science and Technology, Shandong University, Qingdao, China","institution_ids":["https://openalex.org/I80143920"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5109684348"],"corresponding_institution_ids":["https://openalex.org/I177725633","https://openalex.org/I80143920"],"apc_list":null,"apc_paid":null,"fwci":0.7371,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.73022374,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":99},"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":0.9998999834060669,"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":0.9998999834060669,"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":0.9991000294685364,"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/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9969000220298767,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/toolchain","display_name":"Toolchain","score":0.9212098717689514},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7958050966262817},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7493140697479248},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6299484968185425},{"id":"https://openalex.org/keywords/inference-engine","display_name":"Inference engine","score":0.6042032241821289},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.5894678235054016},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5227980017662048},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5226625800132751},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.5040613412857056},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.4786157011985779},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.4559727907180786},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.41802021861076355},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3820043206214905},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3274202346801758},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.28311973810195923},{"id":"https://openalex.org/keywords/wireless","display_name":"Wireless","score":0.13340485095977783},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10018062591552734},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.09893080592155457}],"concepts":[{"id":"https://openalex.org/C2777062904","wikidata":"https://www.wikidata.org/wiki/Q545406","display_name":"Toolchain","level":3,"score":0.9212098717689514},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7958050966262817},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7493140697479248},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6299484968185425},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.6042032241821289},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.5894678235054016},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5227980017662048},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5226625800132751},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.5040613412857056},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.4786157011985779},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4559727907180786},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.41802021861076355},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3820043206214905},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3274202346801758},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28311973810195923},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.13340485095977783},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10018062591552734},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.09893080592155457},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/dac56929.2023.10247793","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac56929.2023.10247793","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 60th ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth","score":0.6600000262260437}],"awards":[],"funders":[{"id":"https://openalex.org/F4320311026","display_name":"Shandong University","ror":"https://ror.org/0207yh398"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W2761757071","https://openalex.org/W2884675507","https://openalex.org/W2895540242","https://openalex.org/W2963701245","https://openalex.org/W2981854066","https://openalex.org/W2997106510","https://openalex.org/W3003306432","https://openalex.org/W3088552745","https://openalex.org/W3104284241","https://openalex.org/W3160369968","https://openalex.org/W3197391509","https://openalex.org/W3216619612","https://openalex.org/W4200195825","https://openalex.org/W4200355694","https://openalex.org/W4220837296"],"related_works":["https://openalex.org/W2013037783","https://openalex.org/W2909413202","https://openalex.org/W4385243142","https://openalex.org/W1999008563","https://openalex.org/W2561644314","https://openalex.org/W2794118724","https://openalex.org/W2912135124","https://openalex.org/W4206450104","https://openalex.org/W2883257033","https://openalex.org/W3082104242"],"abstract_inverted_index":{"The":[0],"Deep":[1],"Learning":[2],"Processor":[3],"(DPU)":[4],"programmable":[5],"engine":[6],"released":[7],"by":[8],"the":[9,19,46,52,134,142,152,162,178],"official":[10],"Xilinx":[11,31,47,179],"Vitis":[12],"AI":[13],"toolchain":[14,48],"has":[15],"become":[16],"one":[17],"of":[18,54,73,93,169],"commercial":[20],"off-the-shelf":[21],"(COTS)":[22],"solutions":[23],"for":[24,82,117],"Convolutional":[25],"Neural":[26],"Networks":[27],"(CNNs)":[28],"inference":[29,62,96,106],"on":[30,177],"FPGAs.":[32],"While":[33],"modern":[34],"FPGA":[35,85,124],"devices":[36],"generally":[37],"have":[38],"enough":[39],"hardware":[40],"resources":[41],"to":[42],"accommodate":[43],"multi-DPUs":[44],"simultaneously,":[45],"currently":[49],"only":[50],"supports":[51],"deployment":[53,72],"multiple":[55,74],"homogeneous":[56,156],"DPUs":[57],"engines":[58,77,101,158],"that":[59,71,90,131],"running":[60],"independent":[61],"tasks":[63],"(task-level":[64],"parallelism).":[65],"In":[66],"this":[67],"work,":[68],"we":[69,88,126],"demonstrate":[70],"heterogeneous":[75,99,136],"DPU":[76,137],"makes":[78],"better":[79],"resource":[80],"efficiency":[81],"a":[83,94,118,128],"given":[84,119],"device.":[86],"Moreover,":[87],"show":[89],"pipelined":[91,148],"execution":[92,143],"CNN":[95,111,120],"task":[97],"over":[98],"multi-DPU":[100,157],"may":[102],"further":[103],"improve":[104],"overall":[105],"throughput":[107],"with":[108,151,155],"carefully":[109],"designed":[110],"layers-to-DPU":[112],"mapping":[113],"and":[114,122,139,147,159,173,184],"scheduling.":[115],"Finally,":[116],"model":[121],"an":[123,166],"device,":[125],"propose":[127],"comprehensive":[129],"framework":[130,164],"automatically":[132],"determines":[133],"optimal":[135],"deployment,":[138],"adaptively":[140],"chooses":[141],"scheme":[144],"between":[145],"task-level":[146],"parallelism.":[149],"Compared":[150],"state-of-the-art":[153],"solution":[154],"network-level":[160],"parallelism,":[161],"proposed":[163],"shows":[165],"average":[167],"improvement":[168],"13%":[170],"(up-to":[171,175],"19%)":[172],"6.6%":[174],"10%)":[176],"Zynq":[180],"UltraScale+":[181],"MPSoC":[182],"ZCU104":[183],"ZCU102":[185],"platforms,":[186],"respectively.":[187]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":1}],"updated_date":"2026-02-25T23:00:34.991745","created_date":"2025-10-10T00:00:00"}