{"id":"https://openalex.org/W3107207569","doi":"https://doi.org/10.1109/icnp49622.2020.9259361","title":"Machine Learning at the Edge: Efficient Utilization of Limited CPU/GPU Resources by Multiplexing","display_name":"Machine Learning at the Edge: Efficient Utilization of Limited CPU/GPU Resources by Multiplexing","publication_year":2020,"publication_date":"2020-10-13","ids":{"openalex":"https://openalex.org/W3107207569","doi":"https://doi.org/10.1109/icnp49622.2020.9259361","mag":"3107207569"},"language":"en","primary_location":{"id":"doi:10.1109/icnp49622.2020.9259361","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icnp49622.2020.9259361","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 IEEE 28th International Conference on Network Protocols (ICNP)","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/A5077061884","display_name":"Aditya Dhakal","orcid":"https://orcid.org/0000-0002-8297-8525"},"institutions":[{"id":"https://openalex.org/I103635307","display_name":"University of California, Riverside","ror":"https://ror.org/03nawhv43","country_code":"US","type":"education","lineage":["https://openalex.org/I103635307"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aditya Dhakal","raw_affiliation_strings":["University of California, Riverside"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California, Riverside","institution_ids":["https://openalex.org/I103635307"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081037214","display_name":"Sameer G. Kulkarni","orcid":"https://orcid.org/0000-0003-4727-6875"},"institutions":[{"id":"https://openalex.org/I27674431","display_name":"Indian Institute of Technology Gandhinagar","ror":"https://ror.org/0036p5w23","country_code":"IN","type":"education","lineage":["https://openalex.org/I27674431"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sameer G Kulkarni","raw_affiliation_strings":["Indian Institute of Technology, Gandhinagar"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Indian Institute of Technology, Gandhinagar","institution_ids":["https://openalex.org/I27674431"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5053712998","display_name":"K. K. Ramakrishnan","orcid":"https://orcid.org/0000-0003-1849-5155"},"institutions":[{"id":"https://openalex.org/I103635307","display_name":"University of California, Riverside","ror":"https://ror.org/03nawhv43","country_code":"US","type":"education","lineage":["https://openalex.org/I103635307"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"K. K. Ramakrishnan","raw_affiliation_strings":["University of California, Riverside"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California, Riverside","institution_ids":["https://openalex.org/I103635307"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.7828,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":{"value":0.86458981,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9994999766349792,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.9994999766349792,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9980000257492065,"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/T13553","display_name":"Age of Information Optimization","score":0.9926999807357788,"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":0.8943240642547607},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.7049113512039185},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.7047091722488403},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.5386177897453308},{"id":"https://openalex.org/keywords/general-purpose-computing-on-graphics-processing-units","display_name":"General-purpose computing on graphics processing units","score":0.5101666450500488},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.5047868490219116},{"id":"https://openalex.org/keywords/remote-direct-memory-access","display_name":"Remote direct memory access","score":0.4464416801929474},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.44281792640686035},{"id":"https://openalex.org/keywords/edge-computing","display_name":"Edge computing","score":0.4314349293708801},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.4306320250034332},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.41180184483528137},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3031710386276245},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.22736528515815735},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.1807934045791626},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.15606585144996643},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.13356688618659973},{"id":"https://openalex.org/keywords/wireless","display_name":"Wireless","score":0.09374570846557617}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8943240642547607},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.7049113512039185},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.7047091722488403},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.5386177897453308},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.5101666450500488},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.5047868490219116},{"id":"https://openalex.org/C130795937","wikidata":"https://www.wikidata.org/wiki/Q2561570","display_name":"Remote direct memory access","level":2,"score":0.4464416801929474},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.44281792640686035},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.4314349293708801},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.4306320250034332},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.41180184483528137},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3031710386276245},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.22736528515815735},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.1807934045791626},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.15606585144996643},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.13356688618659973},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.09374570846557617}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icnp49622.2020.9259361","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icnp49622.2020.9259361","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2020 IEEE 28th International Conference on Network Protocols (ICNP)","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":20,"referenced_works":["https://openalex.org/W753012316","https://openalex.org/W2043926804","https://openalex.org/W2186615578","https://openalex.org/W2465005408","https://openalex.org/W2513383847","https://openalex.org/W2734607546","https://openalex.org/W2792220137","https://openalex.org/W2887289444","https://openalex.org/W2922395136","https://openalex.org/W2931743911","https://openalex.org/W2953384591","https://openalex.org/W2960833983","https://openalex.org/W2963947694","https://openalex.org/W2965289829","https://openalex.org/W2969539885","https://openalex.org/W6622239757","https://openalex.org/W6686509673","https://openalex.org/W6713134421","https://openalex.org/W6754496896","https://openalex.org/W6760481631"],"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/W2370314112","https://openalex.org/W1912958759","https://openalex.org/W2792081825"],"abstract_inverted_index":{"Edge":[0],"clouds":[1],"can":[2,74,131,147,216],"provide":[3],"very":[4],"responsive":[5],"services":[6],"for":[7],"end-user":[8],"devices":[9],"that":[10,65,106,214],"require":[11,55],"more":[12,124],"significant":[13],"compute":[14],"capabilities":[15],"than":[16],"they":[17],"have.":[18],"But":[19],"edge":[20,50],"cloud":[21],"resources":[22],"such":[23,28,144],"as":[24,29,145],"CPUs":[25],"and":[26,33,222],"accelerators":[27],"GPUs":[30,44,83,119],"are":[31,52],"limited":[32],"must":[34],"be":[35,62,75],"shared":[36],"across":[37,45,168],"multiple":[38],"concurrently":[39],"running":[40],"clients.":[41],"However,":[42],"multiplexing":[43],"applications":[46],"is":[47],"challenging.":[48],"Further,":[49],"servers":[51],"likely":[53],"to":[54,61,71,120,177,181,195,204],"considerable":[56],"amounts":[57],"of":[58,81,88,91,118,129,162,230],"streaming":[59],"data":[60,66,179,191],"processed.":[63],"Getting":[64],"from":[67,94,186,192],"the":[68,72,79,86,111,122,135,166,173,193,200],"network":[69,194],"stream":[70],"GPU":[73,95,100,123,130,133,174,209],"a":[76,104,228],"bottleneck,":[77],"limiting":[78],"amount":[80],"work":[82],"do.":[84],"Finally,":[85],"lack":[87],"prompt":[89],"notification":[90],"job":[92],"completion":[93],"also":[96],"results":[97],"in":[98,110,153],"ineffective":[99],"utilization.":[101],"We":[102,114],"propose":[103],"framework":[105,157,171,198],"addresses":[107],"these":[108],"challenges":[109],"following":[112],"manner.":[113],"utilize":[115],"spatial":[116,127,137,160],"sharing":[117,128,138,161],"multiplex":[121],"efficiently.":[125],"While":[126],"increase":[132],"utilization,":[134],"uncontrolled":[136],"currently":[139],"available":[140],"with":[141],"state-of-the-art":[142],"systems":[143],"CUDA-MPS":[146],"cause":[148],"interference":[149,167],"between":[150],"applications,":[151],"resulting":[152],"unpredictable":[154],"latency.":[155],"Our":[156,170,197],"utilizes":[158],"controlled":[159],"GPU,":[163,182],"which":[164],"limits":[165],"applications.":[169],"uses":[172,199],"DMA":[175],"engine":[176],"offload":[178],"transfer":[180],"therefore":[183],"preventing":[184],"CPU":[185],"being":[187],"bottleneck":[188],"while":[189],"transferring":[190],"GPU.":[196],"CUDA":[201],"event":[202],"library":[203],"have":[205],"timely,":[206],"low":[207,218],"overhead":[208],"notifications.":[210],"Preliminary":[211],"experiments":[212],"show":[213],"we":[215],"achieve":[217],"DNN":[219,224],"inference":[220,225],"latency":[221],"improve":[223],"throughput":[226],"by":[227],"factor":[229],"~":[231],"1.4.":[232]},"counts_by_year":[{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":1},{"year":2020,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}