{"id":"https://openalex.org/W4366957282","doi":"https://doi.org/10.1109/tpds.2023.3269530","title":"NIOT: A Novel Inference Optimization of Transformers on Modern CPUs","display_name":"NIOT: A Novel Inference Optimization of Transformers on Modern CPUs","publication_year":2023,"publication_date":"2023-04-24","ids":{"openalex":"https://openalex.org/W4366957282","doi":"https://doi.org/10.1109/tpds.2023.3269530"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2023.3269530","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2023.3269530","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Transactions on Parallel and Distributed Systems","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/A5030460543","display_name":"Zining Zhang","orcid":"https://orcid.org/0000-0002-7992-426X"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":true,"raw_author_name":"Zining Zhang","raw_affiliation_strings":["Neuron Mobility Pte. Ltd., Singapore","School of Computing, National University of Singapore, Singapore","NUS Centre for Trusted Internet and Community, Singapore"],"affiliations":[{"raw_affiliation_string":"Neuron Mobility Pte. Ltd., Singapore","institution_ids":[]},{"raw_affiliation_string":"School of Computing, National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]},{"raw_affiliation_string":"NUS Centre for Trusted Internet and Community, Singapore","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100394256","display_name":"Yao Chen","orcid":"https://orcid.org/0000-0002-5798-2282"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yao Chen","raw_affiliation_strings":["School of Computing, National University of Singapore, Singapore"],"affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039946576","display_name":"Bingsheng He","orcid":"https://orcid.org/0000-0001-8618-4581"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Bingsheng He","raw_affiliation_strings":["School of Computing, National University of Singapore, Singapore"],"affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101923452","display_name":"Zhenjie Zhang","orcid":"https://orcid.org/0000-0001-7011-8889"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Zhenjie Zhang","raw_affiliation_strings":["Neuron Mobility Pte. Ltd., Singapore","School of Computing, National University of Singapore, Singapore"],"affiliations":[{"raw_affiliation_string":"Neuron Mobility Pte. Ltd., Singapore","institution_ids":[]},{"raw_affiliation_string":"School of Computing, National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5030460543"],"corresponding_institution_ids":["https://openalex.org/I165932596"],"apc_list":null,"apc_paid":null,"fwci":1.0832,"has_fulltext":false,"cited_by_count":9,"citation_normalized_percentile":{"value":0.78980542,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":"34","issue":"6","first_page":"1982","last_page":"1995"},"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/T12072","display_name":"Machine Learning and Algorithms","score":0.9973999857902527,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9945999979972839,"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":0.8554720282554626},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7595503330230713},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.58157879114151},{"id":"https://openalex.org/keywords/memory-hierarchy","display_name":"Memory hierarchy","score":0.5292748212814331},{"id":"https://openalex.org/keywords/simd","display_name":"SIMD","score":0.5226224064826965},{"id":"https://openalex.org/keywords/thread","display_name":"Thread (computing)","score":0.5151792764663696},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4837934672832489},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.47607913613319397},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41047632694244385},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.40592771768569946},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3682519495487213},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3329641819000244},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3319935202598572},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.16981744766235352}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8554720282554626},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7595503330230713},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.58157879114151},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.5292748212814331},{"id":"https://openalex.org/C150552126","wikidata":"https://www.wikidata.org/wiki/Q339387","display_name":"SIMD","level":2,"score":0.5226224064826965},{"id":"https://openalex.org/C138101251","wikidata":"https://www.wikidata.org/wiki/Q213092","display_name":"Thread (computing)","level":2,"score":0.5151792764663696},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4837934672832489},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.47607913613319397},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41047632694244385},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40592771768569946},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3682519495487213},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3329641819000244},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3319935202598572},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.16981744766235352},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2023.3269530","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2023.3269530","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Transactions on Parallel and Distributed Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.46000000834465027}],"awards":[],"funders":[{"id":"https://openalex.org/F4320320698","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49"},{"id":"https://openalex.org/F4320320709","display_name":"National Research Foundation Singapore","ror":"https://ror.org/03cpyc314"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":72,"referenced_works":["https://openalex.org/W1551026808","https://openalex.org/W1585509108","https://openalex.org/W2004916460","https://openalex.org/W2108598243","https://openalex.org/W2116176472","https://openalex.org/W2153190325","https://openalex.org/W2169528473","https://openalex.org/W2302330221","https://openalex.org/W2471164860","https://openalex.org/W2804500013","https://openalex.org/W2806900363","https://openalex.org/W2886885214","https://openalex.org/W2892009249","https://openalex.org/W2896457183","https://openalex.org/W2899663614","https://openalex.org/W2903739847","https://openalex.org/W2914926228","https://openalex.org/W2941814890","https://openalex.org/W2961619211","https://openalex.org/W2963748441","https://openalex.org/W2965373594","https://openalex.org/W2971507840","https://openalex.org/W2980282514","https://openalex.org/W3033411150","https://openalex.org/W3081168214","https://openalex.org/W3094502228","https://openalex.org/W3096395190","https://openalex.org/W3099576124","https://openalex.org/W3102446692","https://openalex.org/W3131922516","https://openalex.org/W3137120824","https://openalex.org/W3138516171","https://openalex.org/W3148573243","https://openalex.org/W3154028478","https://openalex.org/W3162629387","https://openalex.org/W3170874841","https://openalex.org/W3172369659","https://openalex.org/W3177265267","https://openalex.org/W3180266648","https://openalex.org/W4210953084","https://openalex.org/W4244917406","https://openalex.org/W4255653201","https://openalex.org/W4280633999","https://openalex.org/W4287748555","https://openalex.org/W4288337707","https://openalex.org/W4295312788","https://openalex.org/W4295838474","https://openalex.org/W4323654151","https://openalex.org/W4385245566","https://openalex.org/W4388295307","https://openalex.org/W6633126254","https://openalex.org/W6635276414","https://openalex.org/W6739901393","https://openalex.org/W6752057402","https://openalex.org/W6753846162","https://openalex.org/W6755207826","https://openalex.org/W6755977528","https://openalex.org/W6761563299","https://openalex.org/W6761628794","https://openalex.org/W6762805369","https://openalex.org/W6766673545","https://openalex.org/W6766978945","https://openalex.org/W6768817161","https://openalex.org/W6771626834","https://openalex.org/W6778823374","https://openalex.org/W6779728309","https://openalex.org/W6780069040","https://openalex.org/W6780226713","https://openalex.org/W6784333009","https://openalex.org/W6788135285","https://openalex.org/W6792275077","https://openalex.org/W6798259571"],"related_works":["https://openalex.org/W2534771569","https://openalex.org/W2037547261","https://openalex.org/W4311812695","https://openalex.org/W2117788426","https://openalex.org/W1993644742","https://openalex.org/W4242015792","https://openalex.org/W2291920536","https://openalex.org/W2164535554","https://openalex.org/W2133317661","https://openalex.org/W1580179249"],"abstract_inverted_index":{"In":[0,68],"the":[1,11,26,35,90,120],"machine":[2,16],"learning":[3,17],"era,":[4],"model":[5],"inference":[6,52,81],"efficiency":[7,88],"is":[8,20,56],"one":[9],"of":[10,37,100,127],"most":[12],"important":[13],"issues":[14],"for":[15,80,89],"systems.":[18],"It":[19],"a":[21,30,75],"major":[22],"challenge":[23],"to":[24,73,85,114,119],"find":[25],"optimal":[27],"configuration":[28],"in":[29,47,59],"huge":[31],"search":[32],"space":[33],"as":[34,66,105,139],"combinations":[36],"kernel":[38],"fusion,":[39],"memory":[40],"tiling,":[41],"and":[42,50,77,107,141],"thread":[43],"allocation":[44],"strategies":[45,117],"result":[46],"highly":[48],"variable":[49],"unpredictable":[51],"performance.":[53],"The":[54],"problem":[55],"particularly":[57],"pronounced":[58],"models":[60,93],"with":[61],"large":[62],"parameter":[63],"matrices":[64],"such":[65,104,138],"Transformers.":[67],"this":[69],"article,":[70],"we":[71],"aim":[72],"develop":[74],"general":[76],"powerful":[78],"framework":[79],"optimization,":[82],"called":[83],"NIOT,":[84],"achieve":[86],"desirable":[87],"prevailing":[91],"Transformer-like":[92],"on":[94],"CPUs.":[95],"To":[96],"take":[97],"full":[98],"advantage":[99],"modern":[101],"CPU":[102],"features":[103],"SIMD":[106],"cache":[108],"hierarchy,":[109],"NIOT":[110,128],"employs":[111],"various":[112],"methods":[113],"provide":[115],"promising":[116],"tailored":[118],"target":[121],"Transformer":[122],"model.":[123],"Our":[124],"C++":[125],"implementation":[126],"shows":[129],"significant":[130],"performance":[131],"improvements":[132],"over":[133],"popular":[134],"well-optimized":[135],"model-serving":[136],"runtimes":[137],"PyTorch":[140],"ONNXRuntime.":[142]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":4}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}