{"id":"https://openalex.org/W4362505433","doi":"https://doi.org/10.1145/3578245.3584726","title":"How Do ML Jobs Fail in Datacenters? Analysis of a Long-Term Dataset from an HPC Cluster","display_name":"How Do ML Jobs Fail in Datacenters? Analysis of a Long-Term Dataset from an HPC Cluster","publication_year":2023,"publication_date":"2023-04-03","ids":{"openalex":"https://openalex.org/W4362505433","doi":"https://doi.org/10.1145/3578245.3584726"},"language":"en","primary_location":{"id":"doi:10.1145/3578245.3584726","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3578245.3584726","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3578245.3584726","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Companion of the 2023 ACM/SPEC International Conference on Performance Engineering","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3578245.3584726","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5081421197","display_name":"Xiaoyu Chu","orcid":"https://orcid.org/0000-0002-0353-7968"},"institutions":[{"id":"https://openalex.org/I865915315","display_name":"Vrije Universiteit Amsterdam","ror":"https://ror.org/008xxew50","country_code":"NL","type":"education","lineage":["https://openalex.org/I865915315"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Xiaoyu Chu","raw_affiliation_strings":["Vrije Universiteit Amsterdam, Amsterdam, Netherlands"],"raw_orcid":"https://orcid.org/0000-0002-0353-7968","affiliations":[{"raw_affiliation_string":"Vrije Universiteit Amsterdam, Amsterdam, Netherlands","institution_ids":["https://openalex.org/I865915315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003364582","display_name":"Sacheendra Talluri","orcid":"https://orcid.org/0000-0003-3461-4919"},"institutions":[{"id":"https://openalex.org/I865915315","display_name":"Vrije Universiteit Amsterdam","ror":"https://ror.org/008xxew50","country_code":"NL","type":"education","lineage":["https://openalex.org/I865915315"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Sacheendra Talluri","raw_affiliation_strings":["Vrije Universiteit Amsterdam, Amsterdam, Netherlands"],"raw_orcid":"https://orcid.org/0000-0003-3461-4919","affiliations":[{"raw_affiliation_string":"Vrije Universiteit Amsterdam, Amsterdam, Netherlands","institution_ids":["https://openalex.org/I865915315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006950257","display_name":"Laurens Versluis","orcid":null},"institutions":[{"id":"https://openalex.org/I865915315","display_name":"Vrije Universiteit Amsterdam","ror":"https://ror.org/008xxew50","country_code":"NL","type":"education","lineage":["https://openalex.org/I865915315"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Laurens Versluis","raw_affiliation_strings":["Vrije Universiteit Amsterdam, Amsterdam, Netherlands"],"raw_orcid":"https://orcid.org/0000-0002-6999-7297","affiliations":[{"raw_affiliation_string":"Vrije Universiteit Amsterdam, Amsterdam, Netherlands","institution_ids":["https://openalex.org/I865915315"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5006986556","display_name":"Alexandru Iosup","orcid":"https://orcid.org/0000-0001-8030-9398"},"institutions":[{"id":"https://openalex.org/I865915315","display_name":"Vrije Universiteit Amsterdam","ror":"https://ror.org/008xxew50","country_code":"NL","type":"education","lineage":["https://openalex.org/I865915315"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Alexandru Iosup","raw_affiliation_strings":["Vrije Universiteit Amsterdam, Amsterdam, Netherlands"],"raw_orcid":"https://orcid.org/0000-0001-8030-9398","affiliations":[{"raw_affiliation_string":"Vrije Universiteit Amsterdam, Amsterdam, Netherlands","institution_ids":["https://openalex.org/I865915315"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I865915315"],"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":"263","last_page":"268"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10101","display_name":"Cloud Computing and Resource Management","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.991100013256073,"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/T12761","display_name":"Data Stream Mining Techniques","score":0.9800000190734863,"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/workload","display_name":"Workload","score":0.8034573793411255},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7513712048530579},{"id":"https://openalex.org/keywords/cluster","display_name":"Cluster (spacecraft)","score":0.5866370797157288},{"id":"https://openalex.org/keywords/supercomputer","display_name":"Supercomputer","score":0.5463103652000427},{"id":"https://openalex.org/keywords/term","display_name":"Term (time)","score":0.4579119086265564},{"id":"https://openalex.org/keywords/resource","display_name":"Resource (disambiguation)","score":0.43787679076194763},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3618951439857483},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3558214604854584},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.2203940749168396},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.09852242469787598}],"concepts":[{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.8034573793411255},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7513712048530579},{"id":"https://openalex.org/C164866538","wikidata":"https://www.wikidata.org/wiki/Q367351","display_name":"Cluster (spacecraft)","level":2,"score":0.5866370797157288},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.5463103652000427},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.4579119086265564},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.43787679076194763},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3618951439857483},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3558214604854584},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.2203940749168396},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.09852242469787598},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1145/3578245.3584726","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3578245.3584726","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3578245.3584726","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Companion of the 2023 ACM/SPEC International Conference on Performance Engineering","raw_type":"proceedings-article"},{"id":"pmh:oai:research.vu.nl:openaire_cris_publications/7940f6a0-1b7f-49f4-bfab-1674ba384bc8","is_oa":true,"landing_page_url":"https://research.vu.nl/en/publications/7940f6a0-1b7f-49f4-bfab-1674ba384bc8","pdf_url":null,"source":{"id":"https://openalex.org/S4306401107","display_name":"VU Research Portal","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I865915315","host_organization_name":"Vrije Universiteit Amsterdam","host_organization_lineage":["https://openalex.org/I865915315"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Chu, X, Talluri, S, Versluis, L & Iosup, A 2023, How Do ML Jobs Fail in Datacenters? Analysis of a Long-Term Dataset from an HPC Cluster. in ICPE 2023 Companinion : Companion of the 2023 ACM/SPEC International Conference on Performance Engineering. Association for Computing Machinery, Inc, pp. 263-268, 14th Annual ACM/SPEC International Conference on Performance Engineering, ICPE 2023, Coimbra, Portugal, 15/04/23. https://doi.org/10.1145/3578245.3584726","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"pmh:oai:research.vu.nl:openaire/7940f6a0-1b7f-49f4-bfab-1674ba384bc8","is_oa":true,"landing_page_url":"https://hdl.handle.net/1871.1/7940f6a0-1b7f-49f4-bfab-1674ba384bc8","pdf_url":null,"source":{"id":"https://openalex.org/S4306401107","display_name":"VU Research Portal","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I865915315","host_organization_name":"Vrije Universiteit Amsterdam","host_organization_lineage":["https://openalex.org/I865915315"],"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":"Chu, X, Talluri, S, Versluis, L & Iosup, A 2023, How Do ML Jobs Fail in Datacenters? Analysis of a Long-Term Dataset from an HPC Cluster. in ICPE 2023 Companinion : Companion of the 2023 ACM/SPEC International Conference on Performance Engineering. Association for Computing Machinery, Inc, pp. 263-268, 14th Annual ACM/SPEC International Conference on Performance Engineering, ICPE 2023, Coimbra, Portugal, 15/04/23. https://doi.org/10.1145/3578245.3584726","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"pmh:oai:research.vu.nl:publications/7940f6a0-1b7f-49f4-bfab-1674ba384bc8","is_oa":true,"landing_page_url":"http://www.scopus.com/inward/citedby.url?scp=85158836930&partnerID=8YFLogxK","pdf_url":null,"source":{"id":"https://openalex.org/S4306401107","display_name":"VU Research Portal","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I865915315","host_organization_name":"Vrije Universiteit Amsterdam","host_organization_lineage":["https://openalex.org/I865915315"],"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":"Chu, X, Talluri, S, Versluis, L & Iosup, A 2023, How Do ML Jobs Fail in Datacenters? Analysis of a Long-Term Dataset from an HPC Cluster. in ICPE 2023 Companinion : Companion of the 2023 ACM/SPEC International Conference on Performance Engineering. Association for Computing Machinery, Inc, pp. 263-268, 14th Annual ACM/SPEC International Conference on Performance Engineering, ICPE 2023, Coimbra, Portugal, 15/04/23. https://doi.org/10.1145/3578245.3584726","raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"doi:10.1145/3578245.3584726","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3578245.3584726","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3578245.3584726","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Companion of the 2023 ACM/SPEC International Conference on Performance Engineering","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth","score":0.6299999952316284}],"awards":[{"id":"https://openalex.org/G7331901853","display_name":null,"funder_award_id":"EU H2020","funder_id":"https://openalex.org/F4320332999","funder_display_name":"Horizon 2020 Framework Programme"}],"funders":[{"id":"https://openalex.org/F4320321800","display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek","ror":"https://ror.org/04jsz6e67"},{"id":"https://openalex.org/F4320322725","display_name":"China Scholarship Council","ror":"https://ror.org/04atp4p48"},{"id":"https://openalex.org/F4320332999","display_name":"Horizon 2020 Framework Programme","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320337144","display_name":"National Supercomputing Center, Korea Institute of Science and Technology Information","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4362505433.pdf","grobid_xml":"https://content.openalex.org/works/W4362505433.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W1555569599","https://openalex.org/W1601664780","https://openalex.org/W1984564341","https://openalex.org/W1984980195","https://openalex.org/W2031513172","https://openalex.org/W2083393867","https://openalex.org/W2142812297","https://openalex.org/W2331481034","https://openalex.org/W2735418241","https://openalex.org/W2767346922","https://openalex.org/W2883852409","https://openalex.org/W2966250553"],"related_works":["https://openalex.org/W2384867379","https://openalex.org/W2329539859","https://openalex.org/W3191490922","https://openalex.org/W2227905990","https://openalex.org/W2765823764","https://openalex.org/W3214280620","https://openalex.org/W2794038527","https://openalex.org/W2151092287","https://openalex.org/W2523801036","https://openalex.org/W2505959877"],"abstract_inverted_index":{"Reliable":[0],"job":[1,53,85,94,146],"execution":[2],"is":[3,34,47,77],"important":[4],"in":[5,87],"High":[6],"Performance":[7],"Computing":[8],"clusters.":[9,89],"Understanding":[10],"the":[11,64,80,97,110],"failure":[12,15,54],"distribution":[13],"and":[14,26,59,92,101,120,140],"pattern":[16,112],"of":[17,74,82,99,113,124],"jobs":[18,131],"helps":[19],"HPC":[20,40,57,88],"cluster":[21],"managers":[22],"design":[23,28],"better":[24],"systems,":[25],"users":[27],"fault":[29],"tolerant":[30],"systems.":[31],"Machine":[32],"learning":[33,52,84,130],"an":[35],"increasingly":[36],"popular":[37],"workload":[38,66],"for":[39],"clusters":[41,68],"are":[42,127],"used":[43,70],"for.":[44,71],"But,":[45],"there":[46],"little":[48],"information":[49],"on":[50,56],"machine":[51,83,129],"characteristics":[55],"clusters,":[58],"how":[60],"they":[61,148],"differ":[62],"from":[63],"previous":[65],"such":[67],"were":[69],"The":[72],"goal":[73],"our":[75,125],"work":[76],"to":[78],"improve":[79],"understanding":[81],"failures":[86],"We":[90,105],"collect":[91],"analyze":[93,106],"data":[95],"spanning":[96],"whole":[98],"2022,":[100],"over":[102],"2~million":[103],"jobs.":[104],"basic":[107],"statistical":[108],"characteristics,":[109],"time":[111],"failures,":[114,119],"resource":[115],"waste":[116,141],"caused":[117],"by":[118],"their":[121],"autocorrelation.":[122],"Some":[123],"findings":[126],"that":[128],"fail":[132],"at":[133],"a":[134],"higher":[135],"rate":[136],"than":[137],"non-ML":[138],"jobs,":[139],"much":[142],"more":[143],"CPU-time":[144],"per":[145],"when":[147],"fail.":[149]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2023-04-05T00:00:00"}