{"id":"https://openalex.org/W4395683288","doi":"https://doi.org/10.1145/3625687.3625797","title":"nnPerf: Demystifying DNN Runtime Inference Latency on Mobile Platforms","display_name":"nnPerf: Demystifying DNN Runtime Inference Latency on Mobile Platforms","publication_year":2023,"publication_date":"2023-11-12","ids":{"openalex":"https://openalex.org/W4395683288","doi":"https://doi.org/10.1145/3625687.3625797"},"language":"en","primary_location":{"id":"doi:10.1145/3625687.3625797","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3625687.3625797","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st ACM Conference on Embedded Networked Sensor Systems","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/A5101833104","display_name":"Haolin Chu","orcid":"https://orcid.org/0009-0000-7184-3150"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haolin Chu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0000-7184-3150","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060249411","display_name":"Xiaolong Zheng","orcid":"https://orcid.org/0000-0001-7950-6773"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaolong Zheng","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-7950-6773","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100322368","display_name":"Liang Liu","orcid":"https://orcid.org/0000-0002-5040-2468"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liang Liu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-5040-2468","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100710713","display_name":"Huad\u00f3ng Ma","orcid":"https://orcid.org/0000-0002-7199-5047"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huadong Ma","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-7199-5047","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I139759216"],"apc_list":null,"apc_paid":null,"fwci":2.525,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.91382207,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"125","last_page":"137"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.991100013256073,"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"}},"topics":[{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.991100013256073,"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"}},{"id":"https://openalex.org/T11986","display_name":"Scientific Computing and Data Management","score":0.9883000254631042,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11614","display_name":"Cloud Data Security Solutions","score":0.9724000096321106,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.834888756275177},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.652916669845581},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5817670226097107},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.39103347063064575},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3634158670902252},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3572142422199249},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.3383778929710388},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.19639161229133606},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.05508735775947571}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.834888756275177},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.652916669845581},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5817670226097107},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.39103347063064575},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3634158670902252},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3572142422199249},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.3383778929710388},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.19639161229133606},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.05508735775947571}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3625687.3625797","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3625687.3625797","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st ACM Conference on Embedded Networked Sensor Systems","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":41,"referenced_works":["https://openalex.org/W1976063779","https://openalex.org/W2741951152","https://openalex.org/W2766975872","https://openalex.org/W2789860971","https://openalex.org/W2804032941","https://openalex.org/W2886851211","https://openalex.org/W2889402930","https://openalex.org/W2897268228","https://openalex.org/W2915478146","https://openalex.org/W2932077855","https://openalex.org/W2935701729","https://openalex.org/W2946948417","https://openalex.org/W2962861284","https://openalex.org/W2963273111","https://openalex.org/W2963363373","https://openalex.org/W2963918968","https://openalex.org/W2964024268","https://openalex.org/W2964081807","https://openalex.org/W2965658867","https://openalex.org/W2980137827","https://openalex.org/W2981758446","https://openalex.org/W2982083293","https://openalex.org/W2984618279","https://openalex.org/W2999905431","https://openalex.org/W3035582633","https://openalex.org/W3037899338","https://openalex.org/W3097841484","https://openalex.org/W3101962329","https://openalex.org/W3139340366","https://openalex.org/W3165698711","https://openalex.org/W3173154111","https://openalex.org/W3173358825","https://openalex.org/W3175471559","https://openalex.org/W3209828932","https://openalex.org/W3210764291","https://openalex.org/W4221024054","https://openalex.org/W4226479682","https://openalex.org/W4387668760","https://openalex.org/W6744307745","https://openalex.org/W6780119413","https://openalex.org/W6780827055"],"related_works":["https://openalex.org/W2055243143","https://openalex.org/W2418291489","https://openalex.org/W3096519538","https://openalex.org/W2744747300","https://openalex.org/W4321636575","https://openalex.org/W4241166160","https://openalex.org/W2068121105","https://openalex.org/W2384826897","https://openalex.org/W3128807919","https://openalex.org/W3176411177"],"abstract_inverted_index":{"We":[0,85,149],"present":[1],"nnPerf,":[2,62],"a":[3,47,152,164],"real-time":[4],"on-device":[5],"profiler":[6],"designed":[7],"to":[8,46,55,80,136,155,161,168],"collect":[9],"and":[10,26,33,40,75,91,96,119,141,146],"analyze":[11],"the":[12,23,36,63,69,77,133],"DNN":[13,31,48],"model":[14,72,78,137,170],"run-time":[15,50,73],"inference":[16,144],"latency":[17,112,145],"on":[18,88,115,129,172],"mobile":[19,56,64,102,130],"platforms.":[20,103],"nnPerf":[21,87,108,128,160],"demystifies":[22],"hidden":[24],"layers":[25],"metrics":[27],"used":[28],"for":[29],"pursuing":[30],"optimizations":[32],"adaptations":[34],"at":[35],"granularity":[37],"of":[38],"operators":[39],"kernels,":[41],"ensuring":[42],"every":[43],"facet":[44],"contributing":[45],"model's":[49],"efficiency":[51,74],"is":[52],"easily":[53,67],"accessible":[54],"developers":[57,65],"via":[58],"well-defined":[59],"APIs.":[60],"With":[61],"can":[66],"identify":[68],"bottleneck":[70],"in":[71],"optimize":[76],"architecture":[79],"meet":[81],"system-level":[82],"objectives":[83],"(SLO).":[84],"implement":[86],"TFLite":[89],"framework":[90],"evaluate":[92],"its":[93],"e2e-,":[94],"operator-,":[95],"kernel-latency":[97],"profiling":[98,113],"accuracy":[99,114],"across":[100],"four":[101],"The":[104],"results":[105],"show":[106,156],"that":[107,126],"achieves":[109],"consistently":[110],"high":[111],"both":[116],"CPU":[117],"(98.12%)":[118],"GPU":[120],"(99.87%).":[121],"Our":[122],"benchmark":[123],"studies":[124],"demonstrate":[125],"running":[127],"devices":[131],"introduces":[132],"minimum":[134],"overhead":[135],"inference,":[138],"with":[139],"0.231%":[140],"0.605%":[142],"extra":[143],"power":[147],"consumption.":[148],"further":[150],"run":[151],"case":[153],"study":[154],"how":[157],"we":[158],"leverage":[159],"migrate":[162],"OFA,":[163],"SOTA":[165],"NAS":[166],"system,":[167],"kernel-oriented":[169],"optimization":[171],"GPUs.":[173]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":3}],"updated_date":"2026-08-08T07:41:36.138363","created_date":"2025-10-10T00:00:00"}