{"id":"https://openalex.org/W3196921579","doi":"https://doi.org/10.1007/s00607-021-00997-9","title":"Using machine learning to model the training scalability of convolutional neural networks on clusters of GPUs","display_name":"Using machine learning to model the training scalability of convolutional neural networks on clusters of GPUs","publication_year":2021,"publication_date":"2021-08-30","ids":{"openalex":"https://openalex.org/W3196921579","doi":"https://doi.org/10.1007/s00607-021-00997-9","mag":"3196921579"},"language":"en","primary_location":{"id":"doi:10.1007/s00607-021-00997-9","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s00607-021-00997-9","pdf_url":"https://link.springer.com/content/pdf/10.1007/s00607-021-00997-9.pdf","source":{"id":"https://openalex.org/S35593046","display_name":"Computing","issn_l":"0010-485X","issn":["0010-485X","1436-5057"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computing","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s00607-021-00997-9.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5090675958","display_name":"Sergio Barrachina","orcid":"https://orcid.org/0000-0003-4642-7467"},"institutions":[{"id":"https://openalex.org/I10902133","display_name":"Universitat Jaume I","ror":"https://ror.org/02ws1xc11","country_code":"ES","type":"education","lineage":["https://openalex.org/I10902133"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Sergio Barrachina","raw_affiliation_strings":["Universitat Jaume I, Castell\u00f3n, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Jaume I, Castell\u00f3n, Spain","institution_ids":["https://openalex.org/I10902133"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033947083","display_name":"Adri\u00e1n Castell\u00f3","orcid":"https://orcid.org/0000-0002-8576-8451"},"institutions":[{"id":"https://openalex.org/I10902133","display_name":"Universitat Jaume I","ror":"https://ror.org/02ws1xc11","country_code":"ES","type":"education","lineage":["https://openalex.org/I10902133"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Adri\u00e1n Castell\u00f3","raw_affiliation_strings":["Universitat Jaume I, Castell\u00f3n, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Jaume I, Castell\u00f3n, Spain","institution_ids":["https://openalex.org/I10902133"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057221923","display_name":"Mar Catal\u00e1n","orcid":"https://orcid.org/0000-0003-4343-2818"},"institutions":[{"id":"https://openalex.org/I10902133","display_name":"Universitat Jaume I","ror":"https://ror.org/02ws1xc11","country_code":"ES","type":"education","lineage":["https://openalex.org/I10902133"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Mar Catal\u00e1n","raw_affiliation_strings":["Universitat Jaume I, Castell\u00f3n, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Jaume I, Castell\u00f3n, Spain","institution_ids":["https://openalex.org/I10902133"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054862021","display_name":"Manuel F. Dolz","orcid":"https://orcid.org/0000-0001-9466-3398"},"institutions":[{"id":"https://openalex.org/I10902133","display_name":"Universitat Jaume I","ror":"https://ror.org/02ws1xc11","country_code":"ES","type":"education","lineage":["https://openalex.org/I10902133"]}],"countries":["ES"],"is_corresponding":true,"raw_author_name":"Manuel F. Dolz","raw_affiliation_strings":["Universitat Jaume I, Castell\u00f3n, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Jaume I, Castell\u00f3n, Spain","institution_ids":["https://openalex.org/I10902133"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5007319120","display_name":"Jose I. Mestre","orcid":"https://orcid.org/0000-0001-5954-4131"},"institutions":[{"id":"https://openalex.org/I10902133","display_name":"Universitat Jaume I","ror":"https://ror.org/02ws1xc11","country_code":"ES","type":"education","lineage":["https://openalex.org/I10902133"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Jose I. Mestre","raw_affiliation_strings":["Universitat Jaume I, Castell\u00f3n, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Jaume I, Castell\u00f3n, Spain","institution_ids":["https://openalex.org/I10902133"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5054862021"],"corresponding_institution_ids":["https://openalex.org/I10902133"],"apc_list":{"value":2290,"currency":"EUR","value_usd":2890},"apc_paid":{"value":2290,"currency":"EUR","value_usd":2890},"fwci":0.3884,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.61173049,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":"105","issue":"5","first_page":"915","last_page":"934"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9995999932289124,"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":0.9995999932289124,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9991999864578247,"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/T12676","display_name":"Machine Learning and ELM","score":0.9983000159263611,"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":0.8362878561019897},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.7814329266548157},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6736782193183899},{"id":"https://openalex.org/keywords/python","display_name":"Python (programming language)","score":0.6716594696044922},{"id":"https://openalex.org/keywords/perceptron","display_name":"Perceptron","score":0.5799304246902466},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5536353588104248},{"id":"https://openalex.org/keywords/node","display_name":"Node (physics)","score":0.4615052044391632},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4547329545021057},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.45288941264152527},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4393768608570099},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.38056063652038574},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3577410578727722},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.10048314929008484},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.0996568500995636}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8362878561019897},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7814329266548157},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6736782193183899},{"id":"https://openalex.org/C519991488","wikidata":"https://www.wikidata.org/wiki/Q28865","display_name":"Python (programming language)","level":2,"score":0.6716594696044922},{"id":"https://openalex.org/C60908668","wikidata":"https://www.wikidata.org/wiki/Q690207","display_name":"Perceptron","level":3,"score":0.5799304246902466},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5536353588104248},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.4615052044391632},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4547329545021057},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.45288941264152527},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4393768608570099},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.38056063652038574},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3577410578727722},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.10048314929008484},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0996568500995636},{"id":"https://openalex.org/C66938386","wikidata":"https://www.wikidata.org/wiki/Q633538","display_name":"Structural engineering","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1007/s00607-021-00997-9","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s00607-021-00997-9","pdf_url":"https://link.springer.com/content/pdf/10.1007/s00607-021-00997-9.pdf","source":{"id":"https://openalex.org/S35593046","display_name":"Computing","issn_l":"0010-485X","issn":["0010-485X","1436-5057"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computing","raw_type":"journal-article"},{"id":"pmh:oai:repositori.uji.es:10234/195009","is_oa":true,"landing_page_url":"http://hdl.handle.net/10234/195009","pdf_url":"http://hdl.handle.net/10234/195009","source":{"id":"https://openalex.org/S4306400303","display_name":"Repositori UJI (Universitat Jaume I)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I10902133","host_organization_name":"Universitat Jaume I","host_organization_lineage":["https://openalex.org/I10902133"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"doi:10.1007/s00607-021-00997-9","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s00607-021-00997-9","pdf_url":"https://link.springer.com/content/pdf/10.1007/s00607-021-00997-9.pdf","source":{"id":"https://openalex.org/S35593046","display_name":"Computing","issn_l":"0010-485X","issn":["0010-485X","1436-5057"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computing","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1117412284","display_name":null,"funder_award_id":"CDEIGENT/2018/014","funder_id":"https://openalex.org/F4320321864","funder_display_name":"Generalitat Valenciana"},{"id":"https://openalex.org/G2810217769","display_name":null,"funder_award_id":"TIN2017-82972-R","funder_id":"https://openalex.org/F4320315062","funder_display_name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades"},{"id":"https://openalex.org/G5221953459","display_name":null,"funder_award_id":"2019/109","funder_id":"https://openalex.org/F4320321864","funder_display_name":"Generalitat Valenciana"},{"id":"https://openalex.org/G6390004333","display_name":null,"funder_award_id":"PROMETEO/2019/109","funder_id":"https://openalex.org/F4320335711","funder_display_name":"Conselleria d'Educaci\u00f3, Investigaci\u00f3, Cultura i Esport"}],"funders":[{"id":"https://openalex.org/F4320315062","display_name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades","ror":null},{"id":"https://openalex.org/F4320321864","display_name":"Generalitat Valenciana","ror":"https://ror.org/0097mvx21"},{"id":"https://openalex.org/F4320335711","display_name":"Conselleria d'Educaci\u00f3, Investigaci\u00f3, Cultura i Esport","ror":"https://ror.org/01w104f98"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3196921579.pdf","grobid_xml":"https://content.openalex.org/works/W3196921579.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W2131613942","https://openalex.org/W2170796499","https://openalex.org/W2286848013","https://openalex.org/W2443267493","https://openalex.org/W2463861691","https://openalex.org/W2604319603","https://openalex.org/W2749988060","https://openalex.org/W2752512710","https://openalex.org/W2783504145","https://openalex.org/W2785452945","https://openalex.org/W2884700152","https://openalex.org/W2892341857","https://openalex.org/W2903557836","https://openalex.org/W2935041335","https://openalex.org/W2955052486","https://openalex.org/W2963459284","https://openalex.org/W2969976451","https://openalex.org/W2972087877","https://openalex.org/W2988070836","https://openalex.org/W3129254793","https://openalex.org/W3158767672","https://openalex.org/W3174716098","https://openalex.org/W6601323341"],"related_works":["https://openalex.org/W2341492732","https://openalex.org/W3187193180","https://openalex.org/W106542691","https://openalex.org/W4287027380","https://openalex.org/W1699080303","https://openalex.org/W4297799326","https://openalex.org/W3116064965","https://openalex.org/W3193760048","https://openalex.org/W4285822516","https://openalex.org/W1486009489"],"abstract_inverted_index":{"Abstract":[0],"In":[1],"this":[2,80,119],"work,":[3],"we":[4,121],"build":[5],"a":[6,130],"general":[7,26],"piece-wise":[8],"model":[9,27,58,81,128],"to":[10],"analyze":[11],"data-parallel":[12],"(DP)":[13],"training":[14,47,75],"costs":[15],"of":[16,23,37,48,50,61,118,125],"convolutional":[17],"neural":[18],"networks":[19],"(CNNs)":[20],"on":[21,30,88],"clusters":[22],"GPUs.":[24],"This":[25],"is":[28],"based":[29],"i)":[31],"multi-layer":[32],"perceptrons":[33],"(MLPs)":[34],"in":[35,45,59,82],"charge":[36,60],"modeling":[38,62],"the":[39,46,51,63,70,85,123,126],"NVIDIA":[40,64],"cuDNN/cuBLAS":[41],"library":[42,132],"kernels":[43],"involved":[44],"some":[49,106],"state-of-the-art":[52],"CNNs;":[53],"and":[54,102,112],"ii)":[55],"an":[56],"analytical":[57],"NCCL":[65],"Allreduce":[66],"collective":[67],"primitive":[68],"using":[69,79],"Ring":[71],"algorithm.":[72],"The":[73],"CNN":[74],"scalability":[76],"study":[77],"performed":[78],"combination":[83],"with":[84],"Roofline":[86],"technique":[87],"varying":[89],"batch":[90],"sizes,":[91],"node":[92,96],"(floating-point)":[93],"arithmetic":[94],"performance,":[95],"memory":[97],"bandwidth,":[98,101],"network":[99],"link":[100],"cluster":[103,113],"dimension":[104],"unveil":[105],"crucial":[107],"bottlenecks":[108],"at":[109],"both":[110],"GPU":[111],"level.":[114],"To":[115],"provide":[116],"evidence":[117],"analysis,":[120],"validate":[122],"accuracy":[124],"proposed":[127],"against":[129],"Python":[131],"for":[133],"distributed":[134],"deep":[135],"learning":[136],"training.":[137]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":1}],"updated_date":"2026-04-10T15:06:20.359241","created_date":"2025-10-10T00:00:00"}
