{"id":"https://openalex.org/W4317038433","doi":"https://doi.org/10.1109/mm.2023.3237548","title":"Failure Tolerant Training With Persistent Memory Disaggregation Over CXL","display_name":"Failure Tolerant Training With Persistent Memory Disaggregation Over CXL","publication_year":2023,"publication_date":"2023-01-17","ids":{"openalex":"https://openalex.org/W4317038433","doi":"https://doi.org/10.1109/mm.2023.3237548"},"language":"en","primary_location":{"id":"doi:10.1109/mm.2023.3237548","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mm.2023.3237548","pdf_url":null,"source":{"id":"https://openalex.org/S59697426","display_name":"IEEE Micro","issn_l":"0272-1732","issn":["0272-1732","1937-4143"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Micro","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/A5068299135","display_name":"Miryeong Kwon","orcid":"https://orcid.org/0000-0002-0313-1319"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Miryeong Kwon","raw_affiliation_strings":["KAIST and Panmnesia, Daejeon, South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST and Panmnesia, Daejeon, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071614011","display_name":"Junhyeok Jang","orcid":"https://orcid.org/0000-0003-3053-396X"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Junhyeok Jang","raw_affiliation_strings":["KAIST and Panmnesia, Daejeon, South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST and Panmnesia, Daejeon, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069131840","display_name":"Hanjin Choi","orcid":"https://orcid.org/0000-0002-3562-9410"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hanjin Choi","raw_affiliation_strings":["KAIST and Panmnesia, Daejeon, South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST and Panmnesia, Daejeon, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100360897","display_name":"Sangwon Lee","orcid":"https://orcid.org/0000-0001-6960-5487"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sangwon Lee","raw_affiliation_strings":["KAIST and Panmnesia, Daejeon, South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST and Panmnesia, Daejeon, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089239312","display_name":"Myoungsoo Jung","orcid":"https://orcid.org/0000-0002-9832-5801"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Myoungsoo Jung","raw_affiliation_strings":["KAIST and Panmnesia, Daejeon, South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST and Panmnesia, Daejeon, South Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5068299135"],"corresponding_institution_ids":["https://openalex.org/I157485424"],"apc_list":null,"apc_paid":null,"fwci":3.3275,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":{"value":0.92358447,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":"43","issue":"2","first_page":"66","last_page":"75"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9988999962806702,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.998199999332428,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9973999857902527,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.9223220348358154},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.6489601135253906},{"id":"https://openalex.org/keywords/fault-tolerance","display_name":"Fault tolerance","score":0.631039559841156},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.5191178321838379},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.43748152256011963},{"id":"https://openalex.org/keywords/memory-hierarchy","display_name":"Memory hierarchy","score":0.4306159019470215},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.4291401207447052},{"id":"https://openalex.org/keywords/graphics-processing-unit","display_name":"Graphics processing unit","score":0.4121740460395813},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3305526673793793},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.2238522469997406}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.9223220348358154},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.6489601135253906},{"id":"https://openalex.org/C63540848","wikidata":"https://www.wikidata.org/wiki/Q3140932","display_name":"Fault tolerance","level":2,"score":0.631039559841156},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.5191178321838379},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.43748152256011963},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.4306159019470215},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4291401207447052},{"id":"https://openalex.org/C2779851693","wikidata":"https://www.wikidata.org/wiki/Q183484","display_name":"Graphics processing unit","level":2,"score":0.4121740460395813},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3305526673793793},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.2238522469997406}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/mm.2023.3237548","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mm.2023.3237548","pdf_url":null,"source":{"id":"https://openalex.org/S59697426","display_name":"IEEE Micro","issn_l":"0272-1732","issn":["0272-1732","1937-4143"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Micro","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.8999999761581421,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":19,"referenced_works":["https://openalex.org/W2794670651","https://openalex.org/W2887208413","https://openalex.org/W2947737663","https://openalex.org/W2979719709","https://openalex.org/W2984020950","https://openalex.org/W3043023836","https://openalex.org/W3043433718","https://openalex.org/W3103991664","https://openalex.org/W3155243801","https://openalex.org/W3158296418","https://openalex.org/W3197720002","https://openalex.org/W3206171855","https://openalex.org/W4280642915","https://openalex.org/W4281634017","https://openalex.org/W4293023368","https://openalex.org/W4296591840","https://openalex.org/W6753818573","https://openalex.org/W6763737044","https://openalex.org/W6794474486"],"related_works":["https://openalex.org/W2285867394","https://openalex.org/W2354144074","https://openalex.org/W1579918296","https://openalex.org/W2126356880","https://openalex.org/W3102446781","https://openalex.org/W2391167130","https://openalex.org/W2092071486","https://openalex.org/W4283067488","https://openalex.org/W2019484008","https://openalex.org/W2027487876"],"abstract_inverted_index":{"This":[0],"article":[1],"proposes":[2],"TrainingCXL":[3,69,113,136],"that":[4,61,119,135],"can":[5,63],"efficiently":[6],"process":[7],"large-scale":[8],"recommendation":[9,98,151],"datasets":[10],"in":[11,56,85],"the":[12,76,94,102,106,121,148],"pool":[13],"of":[14,97,109,124],"disaggregated":[15],"memory":[16,31,58],"while":[17],"making":[18],"training":[19,80,130,139],"fault":[20],"tolerant":[21],"with":[22],"low":[23],"overhead.":[24],"To":[25],"this":[26],"end,":[27],"we":[28,92],"integrate":[29],"persistent":[30],"(PMEM)":[32],"and":[33,72,82,100,127,142],"graphics":[34],"processing":[35],"unit":[36],"(GPU)":[37],"into":[38],"a":[39],"cache-coherent":[40],"domain":[41],"as":[42],"type":[43],"2.":[44],"Enabling":[45],"Compute":[46],"Express":[47],"Link":[48],"(CXL)":[49],"allows":[50],"PMEM":[51,65],"to":[52,147],"be":[53],"directly":[54],"placed":[55],"GPU\u2019s":[57],"hierarchy,":[59],"such":[60],"GPU":[62],"access":[64],"without":[66],"software":[67],"intervention.":[68],"introduces":[70],"computing":[71],"checkpointing":[73,103,117],"logic":[74],"near":[75],"CXL":[77],"controller,":[78],"thereby":[79],"data":[81],"managing":[83],"persistency":[84],"an":[86,115],"active":[87],"manner.":[88],"Considering":[89],"PMEM\u2019s":[90],"vulnerability,":[91],"utilize":[93],"unique":[95],"characteristics":[96],"models":[99],"take":[101],"overhead":[104],"off":[105],"critical":[107],"path":[108],"their":[110],"training.":[111],"Finally,":[112],"employs":[114],"advanced":[116],"technique":[118],"relaxes":[120],"updating":[122],"sequence":[123],"model":[125],"parameters":[126],"embeddings":[128],"across":[129],"batches.":[131],"The":[132],"evaluation":[133],"shows":[134],"achieves":[137],"5.2\u00d7":[138],"performance":[140],"improvement":[141],"76%":[143],"energy":[144],"savings,":[145],"compared":[146],"modern":[149],"PMEM-based":[150],"systems.":[152]},"counts_by_year":[{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":3}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}
