{"id":"https://openalex.org/W4414500386","doi":"https://doi.org/10.1109/aicas64808.2025.11173089","title":"Enhancing Training Efficiency: A Novel Approach to Handling GPU Failures in Large-Scale Distributed System for LLM Training","display_name":"Enhancing Training Efficiency: A Novel Approach to Handling GPU Failures in Large-Scale Distributed System for LLM Training","publication_year":2025,"publication_date":"2025-04-28","ids":{"openalex":"https://openalex.org/W4414500386","doi":"https://doi.org/10.1109/aicas64808.2025.11173089"},"language":"en","primary_location":{"id":"doi:10.1109/aicas64808.2025.11173089","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aicas64808.2025.11173089","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 7th International Conference on Artificial Intelligence Circuits and Systems (AICAS)","raw_type":"proceedings-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/A5081338858","display_name":"Kiran Gunnam","orcid":"https://orcid.org/0000-0001-7219-7234"},"institutions":[{"id":"https://openalex.org/I11912373","display_name":"Micron (United States)","ror":"https://ror.org/02fv52296","country_code":"US","type":"company","lineage":["https://openalex.org/I11912373"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kiran Gunnam","raw_affiliation_strings":["Micron Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Micron Technology","institution_ids":["https://openalex.org/I11912373"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055880305","display_name":"Pranshu Mandal","orcid":null},"institutions":[{"id":"https://openalex.org/I11912373","display_name":"Micron (United States)","ror":"https://ror.org/02fv52296","country_code":"US","type":"company","lineage":["https://openalex.org/I11912373"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pranshu Tapan Mandal","raw_affiliation_strings":["Micron Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Micron Technology","institution_ids":["https://openalex.org/I11912373"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026806172","display_name":"Shivam Khandelwal","orcid":"https://orcid.org/0000-0002-1535-8959"},"institutions":[{"id":"https://openalex.org/I11912373","display_name":"Micron (United States)","ror":"https://ror.org/02fv52296","country_code":"US","type":"company","lineage":["https://openalex.org/I11912373"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shivam Khandelwal","raw_affiliation_strings":["Micron Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Micron Technology","institution_ids":["https://openalex.org/I11912373"]}]},{"author_position":"last","author":{"id":null,"display_name":"Rajesh Bhagwat","orcid":null},"institutions":[{"id":"https://openalex.org/I11912373","display_name":"Micron (United States)","ror":"https://ror.org/02fv52296","country_code":"US","type":"company","lineage":["https://openalex.org/I11912373"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rajesh Bhagwat","raw_affiliation_strings":["Micron Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Micron Technology","institution_ids":["https://openalex.org/I11912373"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.26860279,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9398999810218811,"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":0.9398999810218811,"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/downtime","display_name":"Downtime","score":0.7950000166893005},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5812000036239624},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5670999884605408},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.5260999798774719},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3476000130176544},{"id":"https://openalex.org/keywords/execution-time","display_name":"Execution time","score":0.3361999988555908}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8016999959945679},{"id":"https://openalex.org/C180591934","wikidata":"https://www.wikidata.org/wiki/Q1253369","display_name":"Downtime","level":2,"score":0.7950000166893005},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5812000036239624},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5670999884605408},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5260999798774719},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4593000113964081},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.35850000381469727},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.351500004529953},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3476000130176544},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.3361999988555908},{"id":"https://openalex.org/C2776857766","wikidata":"https://www.wikidata.org/wiki/Q7832987","display_name":"Training system","level":2,"score":0.3345000147819519},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.3181000053882599},{"id":"https://openalex.org/C63540848","wikidata":"https://www.wikidata.org/wiki/Q3140932","display_name":"Fault tolerance","level":2,"score":0.3147999942302704},{"id":"https://openalex.org/C19012869","wikidata":"https://www.wikidata.org/wiki/Q578372","display_name":"Response time","level":2,"score":0.30550000071525574},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.29809999465942383},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2676999866962433},{"id":"https://openalex.org/C3017489831","wikidata":"https://www.wikidata.org/wiki/Q2393193","display_name":"Running time","level":2,"score":0.2535000145435333}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/aicas64808.2025.11173089","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aicas64808.2025.11173089","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 7th International Conference on Artificial Intelligence Circuits and Systems (AICAS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"In":[0],"large-scale":[1],"distributed":[2],"machine":[3],"learning":[4],"systems":[5],"for":[6,113],"LLM":[7],"Training,":[8],"GPU":[9,30,116],"failures":[10,31],"can":[11,54],"significantly":[12],"disrupt":[13],"training":[14,95],"processes,":[15],"often":[16],"necessitating":[17],"restoration":[18],"from":[19,35,115],"prior":[20],"checkpoints.":[21],"This":[22,73],"paper":[23],"proposes":[24],"an":[25],"alternative":[26],"approach":[27,102],"to":[28,71,76],"handle":[29],"by":[32,39],"ignoring":[33],"updates":[34],"the":[36,40,61,78,89,94,108],"batches":[37,46],"processed":[38],"failed":[41],"GPUs":[42],"and":[43,80,91,110],"rescheduling":[44,53],"these":[45],"at":[47,56],"a":[48,57,69],"later":[49],"feasible":[50],"time.":[51],"The":[52],"occur":[55],"subsequent":[58],"time":[59,63,109],"step,":[60,64],"final":[62],"or":[65],"as":[66],"part":[67],"of":[68,93],"prologue":[70],"pre-training.":[72],"method":[74],"aims":[75],"minimize":[77],"downtime":[79],"computational":[81],"overhead":[82],"associated":[83],"with":[84],"checkpoint":[85],"restoration,":[86],"thereby":[87],"enhancing":[88],"efficiency":[90],"robustness":[92],"process.":[96],"Experimental":[97],"results":[98],"demonstrate":[99],"that":[100],"this":[101],"maintains":[103],"model":[104],"accuracy":[105],"while":[106],"reducing":[107],"resources":[111],"required":[112],"recovery":[114],"failures.":[117]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}