{"id":"https://openalex.org/W4205369066","doi":"https://doi.org/10.1109/comsnets53615.2022.9668515","title":"DEFER: Distributed Edge Inference for Deep Neural Networks","display_name":"DEFER: Distributed Edge Inference for Deep Neural Networks","publication_year":2022,"publication_date":"2022-01-04","ids":{"openalex":"https://openalex.org/W4205369066","doi":"https://doi.org/10.1109/comsnets53615.2022.9668515"},"language":"en","primary_location":{"id":"doi:10.1109/comsnets53615.2022.9668515","is_oa":false,"landing_page_url":"https://doi.org/10.1109/comsnets53615.2022.9668515","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 14th International Conference on COMmunication Systems & NETworkS (COMSNETS)","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2201.06769","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5038069138","display_name":"Arjun Parthasarathy","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Arjun Parthasarathy","raw_affiliation_strings":["Crystal Springs Uplands School"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Crystal Springs Uplands School","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5063784062","display_name":"Bhaskar Krishnamachari","orcid":"https://orcid.org/0000-0002-9994-9931"},"institutions":[{"id":"https://openalex.org/I1174212","display_name":"University of Southern California","ror":"https://ror.org/03taz7m60","country_code":"US","type":"education","lineage":["https://openalex.org/I1174212"]},{"id":"https://openalex.org/I2800817003","display_name":"California Southern University","ror":"https://ror.org/058zz0t50","country_code":"US","type":"education","lineage":["https://openalex.org/I2800817003"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bhaskar Krishnamachari","raw_affiliation_strings":["University of Southern California"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Southern California","institution_ids":["https://openalex.org/I2800817003","https://openalex.org/I1174212"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.346,"has_fulltext":false,"cited_by_count":27,"citation_normalized_percentile":{"value":0.89655631,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"749","last_page":"753"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9969000220298767,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.996399998664856,"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/computer-science","display_name":"Computer science","score":0.8300995826721191},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6570513844490051},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.6388660669326782},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5571771264076233},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5551321506500244},{"id":"https://openalex.org/keywords/edge-computing","display_name":"Edge computing","score":0.5378763675689697},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.482481986284256},{"id":"https://openalex.org/keywords/node","display_name":"Node (physics)","score":0.4674535393714905},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.44575053453445435},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.42119699716567993},{"id":"https://openalex.org/keywords/header","display_name":"Header","score":0.4208703637123108},{"id":"https://openalex.org/keywords/energy-consumption","display_name":"Energy consumption","score":0.41790932416915894},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4114900529384613},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.395092248916626},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.3689630925655365},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3324170708656311},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.316919207572937},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.19421660900115967},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.17116019129753113},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.09425705671310425}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8300995826721191},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6570513844490051},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6388660669326782},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5571771264076233},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5551321506500244},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.5378763675689697},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.482481986284256},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.4674535393714905},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.44575053453445435},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.42119699716567993},{"id":"https://openalex.org/C48105269","wikidata":"https://www.wikidata.org/wiki/Q1141160","display_name":"Header","level":2,"score":0.4208703637123108},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.41790932416915894},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4114900529384613},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.395092248916626},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.3689630925655365},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3324170708656311},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.316919207572937},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.19421660900115967},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.17116019129753113},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.09425705671310425},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.0},{"id":"https://openalex.org/C18903297","wikidata":"https://www.wikidata.org/wiki/Q7150","display_name":"Ecology","level":1,"score":0.0},{"id":"https://openalex.org/C66938386","wikidata":"https://www.wikidata.org/wiki/Q633538","display_name":"Structural engineering","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/comsnets53615.2022.9668515","is_oa":false,"landing_page_url":"https://doi.org/10.1109/comsnets53615.2022.9668515","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 14th International Conference on COMmunication Systems & NETworkS (COMSNETS)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2201.06769","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2201.06769","pdf_url":"https://arxiv.org/pdf/2201.06769","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2201.06769","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2201.06769","pdf_url":"https://arxiv.org/pdf/2201.06769","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"score":0.9100000262260437,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W2052440657","https://openalex.org/W2168231600","https://openalex.org/W2194775991","https://openalex.org/W2554777827","https://openalex.org/W2774000609","https://openalex.org/W2896180420","https://openalex.org/W2960833983","https://openalex.org/W2962883027","https://openalex.org/W2980150456","https://openalex.org/W3021654819","https://openalex.org/W3026206397","https://openalex.org/W3048233142","https://openalex.org/W3058843122","https://openalex.org/W3099970223","https://openalex.org/W3103802018","https://openalex.org/W3108787771","https://openalex.org/W4302440534","https://openalex.org/W6684859321","https://openalex.org/W6781915296","https://openalex.org/W6785348223"],"related_works":["https://openalex.org/W4322761281","https://openalex.org/W4238233472","https://openalex.org/W4313526662","https://openalex.org/W3111395152","https://openalex.org/W4312996489","https://openalex.org/W3106131444","https://openalex.org/W3216099748","https://openalex.org/W4205963435","https://openalex.org/W4313463379","https://openalex.org/W3214037210"],"abstract_inverted_index":{"Modern":[0],"machine":[1],"learning":[2,35],"tools":[3],"such":[4,18],"as":[5,19,248],"deep":[6,34,100],"neural":[7,101],"networks":[8,102],"(DNNs)":[9],"are":[10,32,135],"playing":[11],"a":[12,65,92,117,138],"revolutionary":[13],"role":[14],"in":[15,137,235],"many":[16],"fields":[17],"natural":[20],"language":[21],"processing,":[22],"computer":[23],"vision,":[24],"and":[25,46,61,81,125,168,201,220,227,240,244],"the":[26,58,71,149,158,162,178,186,189,224,238,254],"internet":[27],"of":[28,116,192],"things.":[29],"Once":[30],"they":[31],"trained,":[33],"models":[36],"can":[37,69,106],"be":[38,107],"deployed":[39],"on":[40,48,64],"edge":[41,67,96],"computers":[42],"to":[43,121,128,148,157,252],"perform":[44],"classification":[45],"prediction":[47],"real-time":[49],"data":[50,127],"for":[51,55,74,94,170,185],"these":[52],"applications.":[53],"Particularly":[54],"large":[56],"models,":[57],"limited":[59],"computational":[60],"memory":[62],"resources":[63],"single":[66,118,211],"device":[68,212],"become":[70],"throughput":[72,80,191],"bottleneck":[73],"an":[75,249],"inference":[76,126,190],"pipeline.":[77],"To":[78],"increase":[79],"decrease":[82],"per-device":[83],"compute":[84,111,130,133,196],"load,":[85],"we":[86],"present":[87],"DEFER":[88,193,234,247],"(Distributed":[89],"Edge":[90],"inFERence),":[91],"framework":[93,172,251],"distributed":[95],"inference,":[97],"which":[98],"partitions":[99,124],"into":[103],"layers":[104],"that":[105,184],"spread":[108],"across":[109],"multiple":[110],"nodes.":[112,131],"The":[113,132,152],"architecture":[114],"consists":[115],"\u201cdispatcher\u201d":[119],"node":[120,203],"distribute":[122],"DNN":[123],"respective":[129],"nodes":[134,197],"connected":[136],"series":[139],"pattern":[140],"where":[141],"each":[142],"node's":[143],"computed":[144],"result":[145,153],"is":[146,154,198,206],"relayed":[147],"subsequent":[150],"node.":[151],"then":[155],"returned":[156],"Dispatcher.":[159],"We":[160,182,214,231],"quantify":[161],"throughput,":[163],"energy":[164,204,221],"consumption,":[165],"network":[166,175,180,217],"payload,":[167],"overhead":[169],"our":[171],"under":[173],"realistic":[174],"conditions":[176],"using":[177,223,237],"CORE":[179],"emulator.":[181],"find":[183],"ResNet50":[187],"model,":[188],"with":[194],"8":[195],"53%":[199],"higher":[200],"per":[202],"consumption":[205,222],"63":[207],"%":[208],"lower":[209],"than":[210],"inference.":[213],"further":[215],"reduce":[216],"communication":[218],"demands":[219],"ZFP":[225],"serialization":[226],"LZ4":[228],"compression":[229],"algorithms.":[230],"have":[232,245],"implemented":[233],"Python":[236],"TensorFlow":[239],"Keras":[241],"ML":[242],"libraries,":[243],"released":[246],"open-source":[250],"benefit":[253],"research":[255],"community.":[256]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":7},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":4}],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-10T00:00:00"}