{"id":"https://openalex.org/W4205471059","doi":"https://doi.org/10.1109/bigdata52589.2021.9671725","title":"An efficient and flexible inference system for serving heterogeneous ensembles of deep neural networks","display_name":"An efficient and flexible inference system for serving heterogeneous ensembles of deep neural networks","publication_year":2021,"publication_date":"2021-12-15","ids":{"openalex":"https://openalex.org/W4205471059","doi":"https://doi.org/10.1109/bigdata52589.2021.9671725"},"language":"en","primary_location":{"id":"doi:10.1109/bigdata52589.2021.9671725","is_oa":false,"landing_page_url":"https://doi.org/10.1109/bigdata52589.2021.9671725","pdf_url":null,"source":{"id":"https://openalex.org/S4363607718","display_name":"2021 IEEE International Conference on Big Data (Big Data)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE International Conference on Big Data (Big Data)","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2208.14049","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5056690572","display_name":"Pierrick Pochelu","orcid":"https://orcid.org/0000-0002-3525-5033"},"institutions":[{"id":"https://openalex.org/I103084370","display_name":"Total (France)","ror":"https://ror.org/04sk34n56","country_code":"FR","type":"company","lineage":["https://openalex.org/I103084370"]}],"countries":["FR"],"is_corresponding":true,"raw_author_name":"Pierrick Pochelu","raw_affiliation_strings":["TotalEnergies SE, Pau, France"],"affiliations":[{"raw_affiliation_string":"TotalEnergies SE, Pau, France","institution_ids":["https://openalex.org/I103084370"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108528905","display_name":"Serge G. Petiton","orcid":"https://orcid.org/0000-0002-8423-3682"},"institutions":[{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"funder","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I2279609970","display_name":"Universit\u00e9 de Lille","ror":"https://ror.org/02kzqn938","country_code":"FR","type":"education","lineage":["https://openalex.org/I2279609970"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Serge G. Petiton","raw_affiliation_strings":["CNRS, UMR 9189 CRIStAL, Univ. Lille, Lille, France"],"affiliations":[{"raw_affiliation_string":"CNRS, UMR 9189 CRIStAL, Univ. Lille, Lille, France","institution_ids":["https://openalex.org/I2279609970","https://openalex.org/I1294671590"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5061266498","display_name":"Bruno Conche","orcid":null},"institutions":[{"id":"https://openalex.org/I103084370","display_name":"Total (France)","ror":"https://ror.org/04sk34n56","country_code":"FR","type":"company","lineage":["https://openalex.org/I103084370"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Bruno Conche","raw_affiliation_strings":["TotalEnergies SE, Pau, France"],"affiliations":[{"raw_affiliation_string":"TotalEnergies SE, Pau, France","institution_ids":["https://openalex.org/I103084370"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5056690572"],"corresponding_institution_ids":["https://openalex.org/I103084370"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.16139195,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"5225","last_page":"5232"},"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/T12535","display_name":"Machine Learning and Data Classification","score":0.9988999962806702,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9984999895095825,"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":0.8657222986221313},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.733153223991394},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.7255218029022217},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.6247029900550842},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.5763989686965942},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.46906399726867676},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.4557868540287018},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.434100478887558},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.41928520798683167},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.4152848422527313},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.41471874713897705},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.41384217143058777},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38490426540374756},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.34141266345977783},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3367190361022949}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8657222986221313},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.733153223991394},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.7255218029022217},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.6247029900550842},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.5763989686965942},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.46906399726867676},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.4557868540287018},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.434100478887558},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.41928520798683167},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.4152848422527313},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.41471874713897705},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.41384217143058777},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38490426540374756},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34141266345977783},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3367190361022949},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.0},{"id":"https://openalex.org/C120665830","wikidata":"https://www.wikidata.org/wiki/Q14620","display_name":"Optics","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","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},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/bigdata52589.2021.9671725","is_oa":false,"landing_page_url":"https://doi.org/10.1109/bigdata52589.2021.9671725","pdf_url":null,"source":{"id":"https://openalex.org/S4363607718","display_name":"2021 IEEE International Conference on Big Data (Big Data)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE International Conference on Big Data (Big Data)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2208.14049","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2208.14049","pdf_url":"https://arxiv.org/pdf/2208.14049","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:2208.14049","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2208.14049","pdf_url":"https://arxiv.org/pdf/2208.14049","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":[{"display_name":"Industry, innovation and infrastructure","score":0.5799999833106995,"id":"https://metadata.un.org/sdg/9"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W1536680647","https://openalex.org/W2158441306","https://openalex.org/W2901129607","https://openalex.org/W2955715866","https://openalex.org/W2963238274","https://openalex.org/W2964108773","https://openalex.org/W2972016218","https://openalex.org/W2989980710","https://openalex.org/W3003166104","https://openalex.org/W3008535060","https://openalex.org/W3082968028","https://openalex.org/W3127743092","https://openalex.org/W3164450814","https://openalex.org/W4206693140","https://openalex.org/W4287246348","https://openalex.org/W4297808460","https://openalex.org/W6683036876","https://openalex.org/W6730042731","https://openalex.org/W6730956707","https://openalex.org/W6747759466","https://openalex.org/W6765438313","https://openalex.org/W6767542292","https://openalex.org/W6770365600","https://openalex.org/W6792510010"],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2164382479","https://openalex.org/W2146343568","https://openalex.org/W98480971","https://openalex.org/W2150291671","https://openalex.org/W2013643406","https://openalex.org/W2027972911","https://openalex.org/W2157978810","https://openalex.org/W2597809628","https://openalex.org/W1966837078"],"abstract_inverted_index":{"Ensembles":[0],"of":[1,29,49,65,162,187,192],"Deep":[2],"Neural":[3],"Networks":[4],"(DNNs)":[5],"has":[6],"achieved":[7],"qualitative":[8],"predictions":[9],"but":[10],"they":[11],"are":[12],"computing":[13],"and":[14,41,62,92,107,115,135,151,169],"memory":[15,105],"intensive.":[16],"Therefore,":[17],"the":[18,47,104,118,123,136,149,171,183,189,201],"demand":[19],"is":[20,69],"growing":[21],"to":[22,58,81,100,111,130,145,158,198],"make":[23],"them":[24],"answer":[25],"a":[26,54,78,83,98,108,195],"heavy":[27,164],"workload":[28],"requests":[30],"with":[31,60,71,139],"available":[32],"computational":[33],"resources.":[34],"Unlike":[35],"recent":[36],"initiatives":[37],"on":[38,46,127,200],"inference":[39,42,67,124],"servers":[40],"frameworks,":[43],"which":[44],"focus":[45],"prediction":[48],"single":[50,174],"DNNs,":[51],"we":[52,76,121],"propose":[53,77],"new":[55],"software":[56],"layer":[57],"serve":[59,159],"flexibility":[61,150],"efficiency":[63,152],"ensembles":[64],"DNNs.Our":[66],"system":[68,125],"designed":[70],"several":[72],"technical":[73],"innovations.":[74],"First,":[75],"novel":[79],"procedure":[80],"found":[82],"good":[84],"allocation":[85,113],"matrix":[86],"between":[87],"devices":[88,106],"(CPUs":[89],"or":[90],"GPUs)":[91],"DNN":[93,175],"instances.":[94],"It":[95,156,180],"runs":[96],"successively":[97],"worst-fit":[99],"allocate":[101],"DNNs":[102,165,193],"into":[103,166,177],"greedy":[109],"algorithm":[110],"optimize":[112],"settings":[114],"speed":[116],"up":[117,197],"ensemble.":[119],"Second,":[120],"design":[122],"based":[126],"multiple":[128],"processes":[129],"run":[131],"asynchronously:":[132],"batching,":[133],"prediction,":[134],"combination":[137],"rule":[138],"an":[140,160],"efficient":[141],"internal":[142],"communication":[143],"scheme":[144],"avoid":[146],"overhead.Experiments":[147],"show":[148],"under":[153],"extreme":[154],"scenarios:":[155],"successes":[157],"ensemble":[161],"12":[163],"4":[167],"GPUs":[168],"at":[170],"opposite,":[172],"one":[173],"multi-threaded":[176],"16":[178],"GPUs.":[179],"also":[181],"outperforms":[182],"simple":[184],"baseline":[185],"consisting":[186],"optimizing":[188],"batch":[190],"size":[191],"by":[194],"speedup":[196],"2.7X":[199],"image":[202],"classification":[203],"task.":[204]},"counts_by_year":[],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2022-01-25T00:00:00"}