{"id":"https://openalex.org/W4380925941","doi":"https://doi.org/10.1145/3581791.3596842","title":"Boosting DNN Cold Inference on Edge Devices","display_name":"Boosting DNN Cold Inference on Edge Devices","publication_year":2023,"publication_date":"2023-06-16","ids":{"openalex":"https://openalex.org/W4380925941","doi":"https://doi.org/10.1145/3581791.3596842"},"language":"en","primary_location":{"id":"doi:10.1145/3581791.3596842","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3581791.3596842","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 Annual International Conference on Mobile Systems, Applications and Services","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/A5049544852","display_name":"Rongjie Yi","orcid":"https://orcid.org/0009-0003-6040-9596"},"institutions":[{"id":"https://openalex.org/I4392021250","display_name":"State Key Laboratory of Networking and Switching Technology","ror":"https://ror.org/00qtv5q45","country_code":null,"type":"facility","lineage":["https://openalex.org/I139759216","https://openalex.org/I4392021250"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Rongjie Yi","raw_affiliation_strings":["State Key Laboratory of Networking and Switching Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0003-6040-9596","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Networking and Switching Technology, Beijing, China","institution_ids":["https://openalex.org/I4392021250"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101534161","display_name":"Ting Cao","orcid":"https://orcid.org/0000-0002-9107-013X"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ting Cao","raw_affiliation_strings":["Microsoft Research, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-9107-013X","affiliations":[{"raw_affiliation_string":"Microsoft Research, Beijing, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016717153","display_name":"Ao Zhou","orcid":"https://orcid.org/0000-0001-5743-9418"},"institutions":[{"id":"https://openalex.org/I4392021250","display_name":"State Key Laboratory of Networking and Switching Technology","ror":"https://ror.org/00qtv5q45","country_code":null,"type":"facility","lineage":["https://openalex.org/I139759216","https://openalex.org/I4392021250"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ao Zhou","raw_affiliation_strings":["State Key Laboratory of Networking and Switching Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-5743-9418","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Networking and Switching Technology, Beijing, China","institution_ids":["https://openalex.org/I4392021250"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027148228","display_name":"Xiao Ma","orcid":"https://orcid.org/0000-0001-5742-8890"},"institutions":[{"id":"https://openalex.org/I4392021250","display_name":"State Key Laboratory of Networking and Switching Technology","ror":"https://ror.org/00qtv5q45","country_code":null,"type":"facility","lineage":["https://openalex.org/I139759216","https://openalex.org/I4392021250"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiao Ma","raw_affiliation_strings":["State Key Laboratory of Networking and Switching Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-5742-8890","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Networking and Switching Technology, Beijing, China","institution_ids":["https://openalex.org/I4392021250"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054814598","display_name":"Shangguang Wang","orcid":"https://orcid.org/0000-0001-7245-1298"},"institutions":[{"id":"https://openalex.org/I4392021250","display_name":"State Key Laboratory of Networking and Switching Technology","ror":"https://ror.org/00qtv5q45","country_code":null,"type":"facility","lineage":["https://openalex.org/I139759216","https://openalex.org/I4392021250"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shangguang Wang","raw_affiliation_strings":["State Key Laboratory of Networking and Switching Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-7245-1298","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Networking and Switching Technology, Beijing, China","institution_ids":["https://openalex.org/I4392021250"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089690212","display_name":"Mengwei Xu","orcid":"https://orcid.org/0000-0001-6271-6993"},"institutions":[{"id":"https://openalex.org/I4392021250","display_name":"State Key Laboratory of Networking and Switching Technology","ror":"https://ror.org/00qtv5q45","country_code":null,"type":"facility","lineage":["https://openalex.org/I139759216","https://openalex.org/I4392021250"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mengwei Xu","raw_affiliation_strings":["State Key Laboratory of Networking and Switching Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-6271-6993","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Networking and Switching Technology, Beijing, China","institution_ids":["https://openalex.org/I4392021250"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5049544852"],"corresponding_institution_ids":["https://openalex.org/I4392021250"],"apc_list":null,"apc_paid":null,"fwci":1.7662,"has_fulltext":false,"cited_by_count":15,"citation_normalized_percentile":{"value":0.86973715,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"516","last_page":"529"},"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":0.9987999796867371,"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/T12676","display_name":"Machine Learning and ELM","score":0.9987000226974487,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.8624653220176697},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.828009307384491},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7923673391342163},{"id":"https://openalex.org/keywords/boosting","display_name":"Boosting (machine learning)","score":0.6730508804321289},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.5697556138038635},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.564143717288971},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.5077031850814819},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.486010879278183},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.46192947030067444},{"id":"https://openalex.org/keywords/heuristic","display_name":"Heuristic","score":0.43106839060783386},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4017221927642822},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.38308489322662354},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3298490047454834},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.10714298486709595},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.09087687730789185}],"concepts":[{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8624653220176697},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.828009307384491},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7923673391342163},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.6730508804321289},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5697556138038635},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.564143717288971},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.5077031850814819},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.486010879278183},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.46192947030067444},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.43106839060783386},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4017221927642822},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.38308489322662354},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3298490047454834},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.10714298486709595},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.09087687730789185},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3581791.3596842","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3581791.3596842","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 Annual International Conference on Mobile Systems, Applications and Services","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy","score":0.6600000262260437}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":57,"referenced_works":["https://openalex.org/W1131242374","https://openalex.org/W2073601450","https://openalex.org/W2078224158","https://openalex.org/W2097117768","https://openalex.org/W2102346651","https://openalex.org/W2133253683","https://openalex.org/W2133468529","https://openalex.org/W2153000379","https://openalex.org/W2163605009","https://openalex.org/W2172654076","https://openalex.org/W2194775991","https://openalex.org/W2517712916","https://openalex.org/W2525851376","https://openalex.org/W2546536770","https://openalex.org/W2612395950","https://openalex.org/W2623333128","https://openalex.org/W2625157458","https://openalex.org/W2626129225","https://openalex.org/W2860338957","https://openalex.org/W2883780447","https://openalex.org/W2889402930","https://openalex.org/W2897268228","https://openalex.org/W2903650079","https://openalex.org/W2909099954","https://openalex.org/W2913293643","https://openalex.org/W2913340405","https://openalex.org/W2946948417","https://openalex.org/W2950667365","https://openalex.org/W2960833983","https://openalex.org/W2963037989","https://openalex.org/W2963125010","https://openalex.org/W2963163009","https://openalex.org/W2996874060","https://openalex.org/W3006907455","https://openalex.org/W3028307607","https://openalex.org/W3033567383","https://openalex.org/W3049640275","https://openalex.org/W3096484587","https://openalex.org/W3101962329","https://openalex.org/W3102767875","https://openalex.org/W3104494818","https://openalex.org/W3104849992","https://openalex.org/W3136046080","https://openalex.org/W3154608090","https://openalex.org/W3165698711","https://openalex.org/W3209727316","https://openalex.org/W3210764291","https://openalex.org/W3217445637","https://openalex.org/W4236853429","https://openalex.org/W4244652618","https://openalex.org/W4282974189","https://openalex.org/W4288348249","https://openalex.org/W4306178406","https://openalex.org/W4306179717","https://openalex.org/W6632876426","https://openalex.org/W6750378353","https://openalex.org/W6762481227"],"related_works":["https://openalex.org/W2913534066","https://openalex.org/W2970375204","https://openalex.org/W2973622361","https://openalex.org/W3176282186","https://openalex.org/W4387489555","https://openalex.org/W3185576471","https://openalex.org/W4288024917","https://openalex.org/W4293053895","https://openalex.org/W2983364019","https://openalex.org/W2998183476"],"abstract_inverted_index":{"DNNs":[0,20],"are":[1],"ubiquitous":[2],"on":[3,101,109,160,166],"edge":[4,161,167],"devices":[5],"nowadays.":[6],"With":[7],"its":[8,50,140],"increasing":[9],"importance":[10],"and":[11,24,41,49,103,138,163],"use":[12],"cases,":[13],"it's":[14],"not":[15],"likely":[16],"to":[17,38,55,124,151,155],"pack":[18],"all":[19],"into":[21],"device":[22],"memory":[23],"expect":[25],"that":[26,147],"each":[27,88],"inference":[28,67],"has":[29],"been":[30],"warmed":[31],"up.":[32],"Therefore,":[33],"cold":[34,70],"inference,":[35],"the":[36,64,92,98,115,156],"process":[37,95],"read,":[39],"initialize,":[40],"execute":[42],"a":[43,81,121,126,134],"DNN":[44,89,158],"model,":[45],"is":[46,52,73],"becoming":[47],"commonplace":[48],"performance":[51,141],"urgently":[53],"demanded":[54],"be":[56],"optimized.":[57],"To":[58,112],"this":[59],"end,":[60],"we":[61],"present":[62],"NNV12,":[63],"first":[65],"on-device":[66],"engine":[68],"optimizing":[69],"inference.":[71],"NNV12":[72,119,137,148],"built":[74],"atop":[75],"three":[76],"novel":[77],"optimization":[78],"knobs:":[79],"selecting":[80],"proper":[82],"kernel":[83,128],"(i.e.,":[84],"operator":[85],"implementation)":[86],"for":[87],"operator,":[90],"bypassing":[91],"weights":[93,100],"transformation":[94],"by":[96],"caching":[97],"post-transformed":[99],"disk,":[102],"pipelined":[104],"execution":[105],"of":[106,136],"many":[107],"kernels":[108],"asymmetric":[110],"processors.":[111],"tackle":[113],"with":[114],"huge":[116],"search":[117],"space,":[118],"employs":[120],"heuristic-based":[122],"scheme":[123],"obtain":[125],"near-optimal":[127],"scheduling":[129],"plan.":[130],"We":[131],"fully":[132],"implement":[133],"prototype":[135],"evaluate":[139],"across":[142],"extensive":[143],"experiments.":[144],"It":[145],"shows":[146],"achieves":[149],"up":[150],"15.2\u00d7":[152],"speedup":[153,165],"compared":[154],"state-of-the-art":[157],"engines":[159],"CPUs":[162],"401.5\u00d7":[164],"GPUs,":[168],"respectively.":[169]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":6}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}