{"id":"https://openalex.org/W4399596817","doi":"https://doi.org/10.23919/date58400.2024.10546769","title":"Multi-Level Analysis of GPU Utilization in ML Training Workloads","display_name":"Multi-Level Analysis of GPU Utilization in ML Training Workloads","publication_year":2024,"publication_date":"2024-03-25","ids":{"openalex":"https://openalex.org/W4399596817","doi":"https://doi.org/10.23919/date58400.2024.10546769"},"language":"en","primary_location":{"id":"doi:10.23919/date58400.2024.10546769","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date58400.2024.10546769","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 Design, Automation &amp;amp; Test in Europe Conference &amp;amp; Exhibition (DATE)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hal.umontpellier.fr/hal-04523554/document","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022999962","display_name":"Paul Delestrac","orcid":"https://orcid.org/0000-0002-7476-1422"},"institutions":[{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"government","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I19894307","display_name":"Universit\u00e9 de Montpellier","ror":"https://ror.org/051escj72","country_code":"FR","type":"education","lineage":["https://openalex.org/I19894307"]},{"id":"https://openalex.org/I4210101743","display_name":"Laboratoire d'Informatique, de Robotique et de Micro\u00e9lectronique de Montpellier","ror":"https://ror.org/013yean28","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I151295451","https://openalex.org/I19894307","https://openalex.org/I4210101743","https://openalex.org/I4210159245","https://openalex.org/I4405261681"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Paul Delestrac","raw_affiliation_strings":["Univ. Montpellier, CNRS,LIRMM,Montpellier,France"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Univ. Montpellier, CNRS,LIRMM,Montpellier,France","institution_ids":["https://openalex.org/I1294671590","https://openalex.org/I19894307","https://openalex.org/I4210101743"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106514543","display_name":"Debjyoti Battacharjee","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114974","display_name":"IMEC","ror":"https://ror.org/02kcbn207","country_code":"BE","type":"nonprofit","lineage":["https://openalex.org/I4210114974"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Debjyoti Battacharjee","raw_affiliation_strings":["IMEC,Leuven,Belgium"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IMEC,Leuven,Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083744300","display_name":"Simei Yang","orcid":"https://orcid.org/0000-0002-0130-8176"},"institutions":[{"id":"https://openalex.org/I4210114974","display_name":"IMEC","ror":"https://ror.org/02kcbn207","country_code":"BE","type":"nonprofit","lineage":["https://openalex.org/I4210114974"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Simei Yang","raw_affiliation_strings":["IMEC,Leuven,Belgium"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IMEC,Leuven,Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5099110771","display_name":"Diksha Moolchandani","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114974","display_name":"IMEC","ror":"https://ror.org/02kcbn207","country_code":"BE","type":"nonprofit","lineage":["https://openalex.org/I4210114974"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Diksha Moolchandani","raw_affiliation_strings":["IMEC,Leuven,Belgium"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IMEC,Leuven,Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033235745","display_name":"Francky Catthoor","orcid":"https://orcid.org/0000-0002-3599-8515"},"institutions":[{"id":"https://openalex.org/I4210114974","display_name":"IMEC","ror":"https://ror.org/02kcbn207","country_code":"BE","type":"nonprofit","lineage":["https://openalex.org/I4210114974"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Francky Catthoor","raw_affiliation_strings":["IMEC,Leuven,Belgium"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IMEC,Leuven,Belgium","institution_ids":["https://openalex.org/I4210114974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029629432","display_name":"Lionel Torres","orcid":"https://orcid.org/0000-0001-5807-5070"},"institutions":[{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"government","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I19894307","display_name":"Universit\u00e9 de Montpellier","ror":"https://ror.org/051escj72","country_code":"FR","type":"education","lineage":["https://openalex.org/I19894307"]},{"id":"https://openalex.org/I4210101743","display_name":"Laboratoire d'Informatique, de Robotique et de Micro\u00e9lectronique de Montpellier","ror":"https://ror.org/013yean28","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I151295451","https://openalex.org/I19894307","https://openalex.org/I4210101743","https://openalex.org/I4210159245","https://openalex.org/I4405261681"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Lionel Torres","raw_affiliation_strings":["Univ. Montpellier, CNRS,LIRMM,Montpellier,France"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Univ. Montpellier, CNRS,LIRMM,Montpellier,France","institution_ids":["https://openalex.org/I1294671590","https://openalex.org/I19894307","https://openalex.org/I4210101743"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102760981","display_name":"David Novo","orcid":"https://orcid.org/0000-0002-5510-4152"},"institutions":[{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"government","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I19894307","display_name":"Universit\u00e9 de Montpellier","ror":"https://ror.org/051escj72","country_code":"FR","type":"education","lineage":["https://openalex.org/I19894307"]},{"id":"https://openalex.org/I4210101743","display_name":"Laboratoire d'Informatique, de Robotique et de Micro\u00e9lectronique de Montpellier","ror":"https://ror.org/013yean28","country_code":"FR","type":"facility","lineage":["https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I151295451","https://openalex.org/I19894307","https://openalex.org/I4210101743","https://openalex.org/I4210159245","https://openalex.org/I4405261681"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"David Novo","raw_affiliation_strings":["Univ. Montpellier, CNRS,LIRMM,Montpellier,France"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Univ. Montpellier, CNRS,LIRMM,Montpellier,France","institution_ids":["https://openalex.org/I1294671590","https://openalex.org/I19894307","https://openalex.org/I4210101743"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9204999804496765,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9204999804496765,"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":0.7819333672523499},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5608735680580139},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.49083882570266724},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.335537314414978}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7819333672523499},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5608735680580139},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.49083882570266724},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.335537314414978},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.23919/date58400.2024.10546769","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date58400.2024.10546769","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 Design, Automation &amp;amp; Test in Europe Conference &amp;amp; Exhibition (DATE)","raw_type":"proceedings-article"},{"id":"pmh:oai:HAL:hal-04523554v1","is_oa":true,"landing_page_url":"https://hal.umontpellier.fr/hal-04523554","pdf_url":"https://hal.umontpellier.fr/hal-04523554/document","source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"https://www.date-conference.com/","raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"pmh:oai:HAL:hal-04523554v1","is_oa":true,"landing_page_url":"https://hal.umontpellier.fr/hal-04523554","pdf_url":"https://hal.umontpellier.fr/hal-04523554/document","source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"https://www.date-conference.com/","raw_type":"info:eu-repo/semantics/conferenceObject"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320326256","display_name":"Grand \u00c9quipement National De Calcul Intensif","ror":"https://ror.org/0010d1q40"}],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4399596817.pdf"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W230091440","https://openalex.org/W2390279801","https://openalex.org/W2233261550","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2376932109"],"abstract_inverted_index":{"Training":[0],"time":[1],"has":[2],"become":[3],"a":[4,50,77,97],"critical":[5],"bottleneck":[6],"due":[7],"to":[8,18,38,48,117,172],"the":[9,20,28,60,101,111,119,134,137,179,185],"recent":[10],"proliferation":[11],"of":[12,32,53,63,76,100,121,143,184],"large-parameter":[13],"ML":[14,25,33,56,79,128],"models.":[15,26],"GPUs":[16],"continue":[17],"be":[19,88],"prevailing":[21],"architecture":[22],"for":[23],"training":[24,57,80],"However,":[27],"complex":[29],"execution":[30,75,131],"flow":[31],"frameworks":[34],"makes":[35],"it":[36],"difficult":[37],"understand":[39],"GPU":[40,180],"computing":[41,61],"resource":[42],"utilization.":[43],"Our":[44],"main":[45],"goal":[46],"is":[47,175],"provide":[49],"better":[51],"understanding":[52],"how":[54],"efficiently":[55],"workloads":[58],"use":[59],"resources":[62],"modern":[64,161],"GPUs.":[65],"To":[66],"this":[67],"end,":[68],"we":[69,95,109,163],"first":[70],"describe":[71],"an":[72],"ideal":[73],"reference":[74],"GPU-accelerated":[78],"loop":[81],"and":[82,130,136,153,167,188,195],"identify":[83],"relevant":[84],"metrics":[85,112],"that":[86,169],"can":[87],"measured":[89],"using":[90],"existing":[91],"profiling":[92,106],"tools.":[93],"Second,":[94],"produce":[96],"coherent":[98],"integration":[99],"traces":[102],"obtained":[103],"from":[104],"each":[105],"tool.":[107],"Third,":[108],"leverage":[110],"within":[113],"our":[114,157],"integrated":[115],"trace":[116],"analyze":[118],"impact":[120],"different":[122],"software":[123],"optimizations":[124],"(e.g.,":[125],"mixed-precision,":[126],"various":[127],"frameworks,":[129],"modes)":[132],"on":[133,159],"throughput":[135],"associated":[138],"utilization":[139,174,183],"at":[140,178],"multiple":[141],"levels":[142],"hardware":[144],"abstraction":[145],"(i.e.,":[146],"whole":[147],"GPU,":[148],"SM":[149],"subpartitions,":[150],"issue":[151,186],"slots,":[152],"tensor":[154,189],"cores).":[155],"In":[156],"results":[158],"two":[160],"GPUs,":[162],"present":[164],"seven":[165],"takeaways":[166],"show":[168],"although":[170],"close":[171],"100%":[173],"generally":[176],"achieved":[177],"level,":[181],"average":[182],"slots":[187],"cores":[190],"always":[191],"remains":[192],"below":[193],"50%":[194],"5.2%,":[196],"respectively.":[197]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-10-10T00:00:00"}
