{"id":"https://openalex.org/W4386117806","doi":"https://doi.org/10.14778/3611479.3611514","title":"Efficient Fault Tolerance for Recommendation Model Training via Erasure Coding","display_name":"Efficient Fault Tolerance for Recommendation Model Training via Erasure Coding","publication_year":2023,"publication_date":"2023-07-01","ids":{"openalex":"https://openalex.org/W4386117806","doi":"https://doi.org/10.14778/3611479.3611514"},"language":"en","primary_location":{"id":"doi:10.14778/3611479.3611514","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3611479.3611514","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the VLDB Endowment","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100425849","display_name":"Tianyu Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tianyu Zhang","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012816805","display_name":"Kaige Liu","orcid":"https://orcid.org/0000-0002-7145-7353"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kaige Liu","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034516607","display_name":"Jack Kosaian","orcid":"https://orcid.org/0000-0001-8812-7847"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jack Kosaian","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034227509","display_name":"Juncheng Yang","orcid":"https://orcid.org/0000-0002-0412-1139"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Juncheng Yang","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5022601441","display_name":"Rashmi C Vinayak","orcid":null},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rashmi Vinayak","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA","institution_ids":["https://openalex.org/I74973139"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I74973139"],"apc_list":null,"apc_paid":null,"fwci":1.5792,"has_fulltext":false,"cited_by_count":9,"citation_normalized_percentile":{"value":0.84151724,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":"16","issue":"11","first_page":"3137","last_page":"3150"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9980000257492065,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9980000257492065,"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/T10772","display_name":"Distributed systems and fault tolerance","score":0.9976000189781189,"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/T11478","display_name":"Caching and Content Delivery","score":0.9940999746322632,"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":0.8346754312515259},{"id":"https://openalex.org/keywords/fault-tolerance","display_name":"Fault tolerance","score":0.7149595022201538},{"id":"https://openalex.org/keywords/erasure-code","display_name":"Erasure code","score":0.6460673213005066},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.6087778210639954},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.5362986922264099},{"id":"https://openalex.org/keywords/coding","display_name":"Coding (social sciences)","score":0.4630349576473236},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.43182215094566345},{"id":"https://openalex.org/keywords/erasure","display_name":"Erasure","score":0.42998769879341125},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.343014121055603},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.24381014704704285},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.17705506086349487},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.14468857645988464},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.119620680809021}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8346754312515259},{"id":"https://openalex.org/C63540848","wikidata":"https://www.wikidata.org/wiki/Q3140932","display_name":"Fault tolerance","level":2,"score":0.7149595022201538},{"id":"https://openalex.org/C137529215","wikidata":"https://www.wikidata.org/wiki/Q5385031","display_name":"Erasure code","level":3,"score":0.6460673213005066},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.6087778210639954},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5362986922264099},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.4630349576473236},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.43182215094566345},{"id":"https://openalex.org/C2778790127","wikidata":"https://www.wikidata.org/wiki/Q484885","display_name":"Erasure","level":2,"score":0.42998769879341125},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.343014121055603},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.24381014704704285},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.17705506086349487},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.14468857645988464},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.119620680809021},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"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/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.14778/3611479.3611514","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3611479.3611514","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the VLDB Endowment","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":37,"referenced_works":["https://openalex.org/W95608104","https://openalex.org/W1522301498","https://openalex.org/W1984564341","https://openalex.org/W1993143336","https://openalex.org/W2039915451","https://openalex.org/W2146502635","https://openalex.org/W2268702383","https://openalex.org/W2340222647","https://openalex.org/W2512971201","https://openalex.org/W2577787132","https://openalex.org/W2604454907","https://openalex.org/W2741736887","https://openalex.org/W2806273311","https://openalex.org/W2898046929","https://openalex.org/W2900810680","https://openalex.org/W2902687529","https://openalex.org/W2949925098","https://openalex.org/W2984020950","https://openalex.org/W2996471668","https://openalex.org/W2999012726","https://openalex.org/W3016842236","https://openalex.org/W3021124033","https://openalex.org/W3095776306","https://openalex.org/W3100515187","https://openalex.org/W3103649678","https://openalex.org/W3115029474","https://openalex.org/W3121395088","https://openalex.org/W3155243801","https://openalex.org/W3158146252","https://openalex.org/W3158296418","https://openalex.org/W3197870239","https://openalex.org/W3200211247","https://openalex.org/W4214658871","https://openalex.org/W4248578633","https://openalex.org/W6713134421","https://openalex.org/W6764791836","https://openalex.org/W6779748563"],"related_works":["https://openalex.org/W2137028263","https://openalex.org/W2619878113","https://openalex.org/W2982181605","https://openalex.org/W2031335939","https://openalex.org/W1947543223","https://openalex.org/W4287065415","https://openalex.org/W1694938350","https://openalex.org/W2744779861","https://openalex.org/W1985967368","https://openalex.org/W2130518071"],"abstract_inverted_index":{"Deep-learning-based":[0],"recommendation":[1],"models":[2],"(DLRMs)":[3],"are":[4,56,120],"widely":[5],"deployed":[6],"to":[7,13,27,33,66,69,110,122,178,202,213,219],"serve":[8],"personalized":[9],"content.":[10],"In":[11],"addition":[12],"using":[14],"neural":[15],"networks,":[16],"DLRMs":[17,210],"have":[18],"large,":[19],"sparsely-accessed":[20],"embedding":[21,38],"tables,":[22,39],"which":[23,119],"map":[24],"categorical":[25],"features":[26],"a":[28,136,158,229],"learned":[29],"dense":[30],"representation.":[31],"Due":[32],"the":[34,46,76,150,184,187],"large":[35,60,209],"sizes":[36],"of":[37,48,52,153,186],"DLRM":[40,105,131,137,154,167,198],"training":[41,68,138,177,199,224],"is":[42,75,108],"typically":[43],"distributed":[44],"across":[45],"memory":[47],"tens":[49],"or":[50],"hundreds":[51],"nodes.":[53],"Node":[54],"failures":[55],"common":[57],"in":[58,83,130,232],"such":[59],"systems":[61],"and":[62,95,164,170,175,222],"must":[63,236],"be":[64],"mitigated":[65],"enable":[67],"complete":[70],"within":[71],"production":[72],"deadlines.":[73],"Checkpointing":[74],"primary":[77],"approach":[78,160],"used":[79],"for":[80,116,126],"fault":[81,128,143],"tolerance":[82,129,144],"these":[84,101],"systems,":[85],"but":[86],"incurs":[87],"significant":[88],"time":[89],"overhead":[90,115,207],"both":[91],"during":[92,225],"normal":[93],"operation":[94],"when":[96],"recovering":[97],"from":[98,216],"failures.":[99],"As":[100],"overheads":[102],"increase":[103],"with":[104,149,227],"size,":[106],"checkpointing":[107,235],"slated":[109],"become":[111],"an":[112,195],"even":[113],"larger":[114],"future":[117],"DLRMs,":[118],"expected":[121],"grow.":[123],"This":[124],"calls":[125],"rethinking":[127],"training.":[132,155],"We":[133,190],"present":[134],"ECRec,":[135],"system":[139],"that":[140],"achieves":[141],"efficient":[142],"by":[145,211],"coupling":[146],"erasure":[147,162],"coding":[148,163],"unique":[151],"characteristics":[152],"ECRec":[156,192,204],"takes":[157],"hybrid":[159],"between":[161],"replicating":[165],"different":[166],"parameters,":[168,174],"correctly":[169],"efficiently":[171],"updates":[172],"redundant":[173],"enables":[176],"proceed":[179],"without":[180],"pauses,":[181],"while":[182],"maintaining":[183],"consistency":[185],"recovered":[188],"parameters.":[189],"implement":[191],"atop":[193],"XDL,":[194],"open-source,":[196],"industrial-scale":[197],"system.":[200],"Compared":[201],"checkpointing,":[203],"reduces":[205],"training-time":[206],"on":[208],"up":[212,218],"66%,":[214],"recovers":[215],"failure":[217],"9.8\u00d7":[220],"faster,":[221],"continues":[223],"recovery":[226],"only":[228],"7--13%":[230],"drop":[231],"throughput":[233],"(whereas":[234],"pause).":[237]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":1}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}