{"id":"https://openalex.org/W3080950018","doi":"https://doi.org/10.1145/3409963.3410493","title":"Profiling and optimizing deep learning inference on mobile GPUs","display_name":"Profiling and optimizing deep learning inference on mobile GPUs","publication_year":2020,"publication_date":"2020-08-19","ids":{"openalex":"https://openalex.org/W3080950018","doi":"https://doi.org/10.1145/3409963.3410493","mag":"3080950018"},"language":"en","primary_location":{"id":"doi:10.1145/3409963.3410493","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3409963.3410493","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 11th ACM SIGOPS Asia-Pacific Workshop on Systems","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/A5101989691","display_name":"Shiqi Jiang","orcid":"https://orcid.org/0009-0009-4180-2804"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":true,"raw_author_name":"Shiqi Jiang","raw_affiliation_strings":["Microsoft Research"],"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060817435","display_name":"Lihao Ran","orcid":null},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Lihao Ran","raw_affiliation_strings":["Microsoft Research"],"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I4210164937"]}]},{"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/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Ting Cao","raw_affiliation_strings":["Microsoft Research"],"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102867598","display_name":"Yusen Xu","orcid":"https://orcid.org/0000-0002-8542-2442"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yusen Xu","raw_affiliation_strings":["Microsoft Research"],"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I4210164937"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102880548","display_name":"Yunxin Liu","orcid":"https://orcid.org/0000-0001-7352-8955"},"institutions":[{"id":"https://openalex.org/I4210164937","display_name":"Microsoft Research (United Kingdom)","ror":"https://ror.org/05k87vq12","country_code":"GB","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210164937"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yunxin Liu","raw_affiliation_strings":["Microsoft Research"],"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I4210164937"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5101989691"],"corresponding_institution_ids":["https://openalex.org/I4210164937"],"apc_list":null,"apc_paid":null,"fwci":1.4055,"has_fulltext":false,"cited_by_count":18,"citation_normalized_percentile":{"value":0.81023968,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"75","last_page":"81"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9994999766349792,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9994999766349792,"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"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9993000030517578,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9972000122070312,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8527635335922241},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7720093727111816},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.6305184960365295},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.6062776446342468},{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.597943127155304},{"id":"https://openalex.org/keywords/profiling","display_name":"Profiling (computer programming)","score":0.5707182884216309},{"id":"https://openalex.org/keywords/mobile-device","display_name":"Mobile device","score":0.5604725480079651},{"id":"https://openalex.org/keywords/partition","display_name":"Partition (number theory)","score":0.5314582586288452},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5008180141448975},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.48254865407943726},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.4497690498828888},{"id":"https://openalex.org/keywords/mobile-computing","display_name":"Mobile computing","score":0.4475466012954712},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3675546646118164},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.35501331090927124},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3524070084095001},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3300006091594696},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.1385248601436615},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.11071589589118958}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8527635335922241},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7720093727111816},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.6305184960365295},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.6062776446342468},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.597943127155304},{"id":"https://openalex.org/C187191949","wikidata":"https://www.wikidata.org/wiki/Q1138496","display_name":"Profiling (computer programming)","level":2,"score":0.5707182884216309},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.5604725480079651},{"id":"https://openalex.org/C42812","wikidata":"https://www.wikidata.org/wiki/Q1082910","display_name":"Partition (number theory)","level":2,"score":0.5314582586288452},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5008180141448975},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.48254865407943726},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.4497690498828888},{"id":"https://openalex.org/C144543869","wikidata":"https://www.wikidata.org/wiki/Q2738570","display_name":"Mobile computing","level":2,"score":0.4475466012954712},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3675546646118164},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.35501331090927124},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3524070084095001},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3300006091594696},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.1385248601436615},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.11071589589118958},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","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:10.1145/3409963.3410493","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3409963.3410493","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 11th ACM SIGOPS Asia-Pacific Workshop on 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":11,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W1790519466","https://openalex.org/W2183341477","https://openalex.org/W2194775991","https://openalex.org/W2575891491","https://openalex.org/W2772161496","https://openalex.org/W2796649226","https://openalex.org/W2963446712","https://openalex.org/W2964081807","https://openalex.org/W2964279072","https://openalex.org/W3105204543"],"related_works":["https://openalex.org/W2280422768","https://openalex.org/W3143197806","https://openalex.org/W4252555497","https://openalex.org/W3121175838","https://openalex.org/W3016293053","https://openalex.org/W2401723157","https://openalex.org/W2952904874","https://openalex.org/W324626582","https://openalex.org/W4389302559","https://openalex.org/W1690653314"],"abstract_inverted_index":{"Mobile":[0],"GPU,":[1],"as":[2],"the":[3,15,27,43,56,87],"ubiquitous":[4],"computing":[5],"hardware":[6],"on":[7,26,77,94],"almost":[8],"every":[9],"smartphone,":[10],"is":[11,55],"being":[12],"exploited":[13],"for":[14],"deep":[16],"learning":[17],"inference.":[18],"In":[19],"this":[20],"paper,":[21],"we":[22,65],"present":[23],"our":[24,84],"measurements":[25],"inference":[28,88],"performance":[29,74],"with":[30],"mobile":[31,37,78],"GPUs.":[32],"Our":[33],"observations":[34],"suggest":[35],"that":[36,50,83],"GPUs":[38],"are":[39],"underutilized.":[40],"We":[41],"study":[42],"inefficient":[44],"issue":[45],"in":[46],"depth":[47],"and":[48,75,97],"find":[49],"one":[51],"of":[52,59],"root":[53],"causes":[54],"improper":[57],"partition":[58],"compute":[60],"workload.":[61],"To":[62],"solve":[63],"this,":[64],"propose":[66],"a":[67],"heuristics-based":[68],"workload":[69],"partitioning":[70],"approach,":[71],"considering":[72],"both":[73],"overheads":[76],"devices.":[79],"Evaluation":[80],"results":[81],"show":[82],"approach":[85],"reduces":[86],"latency":[89],"by":[90],"up":[91],"to":[92],"32.8%":[93],"various":[95],"devices":[96],"neural":[98],"networks.":[99]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":2}],"updated_date":"2026-03-13T16:22:10.518609","created_date":"2025-10-10T00:00:00"}