{"id":"https://openalex.org/W3169512402","doi":"https://doi.org/10.1109/les.2021.3087707","title":"Deep Learning Inference Parallelization on Heterogeneous Processors With TensorRT","display_name":"Deep Learning Inference Parallelization on Heterogeneous Processors With TensorRT","publication_year":2021,"publication_date":"2021-06-09","ids":{"openalex":"https://openalex.org/W3169512402","doi":"https://doi.org/10.1109/les.2021.3087707","mag":"3169512402"},"language":"en","primary_location":{"id":"doi:10.1109/les.2021.3087707","is_oa":false,"landing_page_url":"https://doi.org/10.1109/les.2021.3087707","pdf_url":null,"source":{"id":"https://openalex.org/S22443479","display_name":"IEEE Embedded Systems Letters","issn_l":"1943-0663","issn":["1943-0663","1943-0671"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Embedded Systems Letters","raw_type":"journal-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/A5028646513","display_name":"Eunjin Jeong","orcid":"https://orcid.org/0000-0002-9585-3369"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"EunJin Jeong","raw_affiliation_strings":["Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea"],"raw_orcid":"https://orcid.org/0000-0002-9585-3369","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058275238","display_name":"Jangryul Kim","orcid":"https://orcid.org/0000-0001-6764-9419"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jangryul Kim","raw_affiliation_strings":["Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea"],"raw_orcid":"https://orcid.org/0000-0001-6764-9419","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025324379","display_name":"Samnieng Tan","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Samnieng Tan","raw_affiliation_strings":["Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100740650","display_name":"Jaeseong Lee","orcid":"https://orcid.org/0000-0002-3311-3891"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jaeseong Lee","raw_affiliation_strings":["Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea"],"raw_orcid":"https://orcid.org/0000-0002-3311-3891","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5029793438","display_name":"Soonhoi Ha","orcid":"https://orcid.org/0000-0001-7472-9142"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Soonhoi Ha","raw_affiliation_strings":["Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea"],"raw_orcid":"https://orcid.org/0000-0001-7472-9142","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":6.5007,"has_fulltext":false,"cited_by_count":89,"citation_normalized_percentile":{"value":0.97612942,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"biblio":{"volume":"14","issue":"1","first_page":"15","last_page":"18"},"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":0.9997000098228455,"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":0.9993000030517578,"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":0.895977258682251},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7944587469100952},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6624845266342163},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.6234717965126038},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.6138606667518616},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.49980998039245605},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4693366587162018},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.4352441132068634},{"id":"https://openalex.org/keywords/general-purpose-computing-on-graphics-processing-units","display_name":"General-purpose computing on graphics processing units","score":0.42590227723121643},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.4243369698524475},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3284812569618225},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.1466747224330902},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.10557487607002258},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.09910151362419128}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.895977258682251},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7944587469100952},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6624845266342163},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.6234717965126038},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.6138606667518616},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.49980998039245605},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4693366587162018},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.4352441132068634},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.42590227723121643},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4243369698524475},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3284812569618225},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.1466747224330902},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.10557487607002258},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.09910151362419128},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/les.2021.3087707","is_oa":false,"landing_page_url":"https://doi.org/10.1109/les.2021.3087707","pdf_url":null,"source":{"id":"https://openalex.org/S22443479","display_name":"IEEE Embedded Systems Letters","issn_l":"1943-0663","issn":["1943-0663","1943-0671"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Embedded Systems Letters","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G7859421288","display_name":null,"funder_award_id":"NRF-2019R1A2B5B02069406","funder_id":"https://openalex.org/F4320322030","funder_display_name":"Ministry of Science, ICT and Future Planning"}],"funders":[{"id":"https://openalex.org/F4320322030","display_name":"Ministry of Science, ICT and Future Planning","ror":"https://ror.org/032e49973"},{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W2525951180","https://openalex.org/W2570343428","https://openalex.org/W2798998154","https://openalex.org/W2885010062","https://openalex.org/W2922395136","https://openalex.org/W3010466514","https://openalex.org/W3014810041","https://openalex.org/W3018757597","https://openalex.org/W3042011474","https://openalex.org/W3092289183","https://openalex.org/W3095320840","https://openalex.org/W4236853429","https://openalex.org/W4293584584","https://openalex.org/W6765718025"],"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/W2546223573","https://openalex.org/W2370314112","https://openalex.org/W1912958759"],"abstract_inverted_index":{"As":[0],"deep":[1],"learning":[2],"(DL)":[3],"inference":[4,69,125],"applications":[5],"are":[6],"increasing,":[7],"an":[8],"embedded":[9],"device":[10],"tends":[11],"to":[12,20,90],"equip":[13],"neural":[14],"processing":[15,75],"units":[16],"(NPUs)":[17],"in":[18],"addition":[19],"a":[21,24,73,87,95],"CPU":[22],"and":[23,28,49,55,102],"GPU.":[25],"For":[26],"fast":[27],"efficient":[29],"development":[30,40],"of":[31,94,108],"DL":[32,59,63,97],"applications,":[33],"TensorRT":[34,65],"is":[35,70],"provided":[36],"as":[37],"the":[38,43,68,92,123],"software":[39],"kit":[41],"for":[42,58],"NVIDIA":[44],"hardware":[45],"platform,":[46],"including":[47],"optimizer":[48],"runtime":[50],"that":[51,67],"delivers":[52],"low":[53],"latency":[54],"high":[56],"throughput":[57,93,120],"inference.":[60],"Like":[61],"most":[62],"frameworks,":[64],"assumes":[66],"executed":[71],"on":[72,110],"single":[74,96],"element,":[76],"GPU,":[77],"or":[78],"NPU,":[79],"not":[80],"both.":[81],"In":[82],"this":[83],"letter,":[84],"we":[85,116],"propose":[86],"parallelization":[88],"methodology":[89],"maximize":[91],"application":[98],"using":[99,126],"both":[100],"GPU":[101,127],"NPU":[103],"by":[104],"exploiting":[105],"various":[106],"types":[107],"parallelism":[109],"TensorRT.":[111],"With":[112],"six":[113],"real-life":[114],"benchmarks,":[115],"could":[117],"achieve":[118],"81%\u2013391%":[119],"improvement":[121],"over":[122],"baseline":[124],"only.":[128]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":19},{"year":2024,"cited_by_count":23},{"year":2023,"cited_by_count":21},{"year":2022,"cited_by_count":19},{"year":2021,"cited_by_count":4}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}