{"id":"https://openalex.org/W4414494300","doi":"https://doi.org/10.1109/aicas64808.2025.11173088","title":"Addition of CXL memory pool to increase LLM training check-pointing frequency","display_name":"Addition of CXL memory pool to increase LLM training check-pointing frequency","publication_year":2025,"publication_date":"2025-04-28","ids":{"openalex":"https://openalex.org/W4414494300","doi":"https://doi.org/10.1109/aicas64808.2025.11173088"},"language":"en","primary_location":{"id":"doi:10.1109/aicas64808.2025.11173088","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aicas64808.2025.11173088","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/A5119725238","display_name":"Prasad Chandrashekara","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":"Prasad Chandrashekara","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/A5119725239","display_name":"Murali Kothandapani","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":"Murali Kothandapani","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":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"]}]},{"author_position":"last","author":{"id":null,"display_name":"Priyanshi Jain","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":"Priyanshi Jain","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":5,"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":0.12198735,"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":"4"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.6011999845504761,"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"}},"topics":[{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.6011999845504761,"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/dram","display_name":"Dram","score":0.7473000288009644},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.6068000197410583},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.4514000117778778},{"id":"https://openalex.org/keywords/work","display_name":"Work (physics)","score":0.39489999413490295},{"id":"https://openalex.org/keywords/node","display_name":"Node (physics)","score":0.37599998712539673},{"id":"https://openalex.org/keywords/power","display_name":"Power (physics)","score":0.3619999885559082},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.3497999906539917}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8246999979019165},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.7473000288009644},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.6068000197410583},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5303999781608582},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.4514000117778778},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.39489999413490295},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.37599998712539673},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.3619999885559082},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.3497999906539917},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3239000141620636},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.323199987411499},{"id":"https://openalex.org/C49154492","wikidata":"https://www.wikidata.org/wiki/Q5300","display_name":"Central processing unit","level":2,"score":0.29589998722076416},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.29429998993873596},{"id":"https://openalex.org/C854659","wikidata":"https://www.wikidata.org/wiki/Q1859284","display_name":"Message passing","level":2,"score":0.2838999927043915},{"id":"https://openalex.org/C193519340","wikidata":"https://www.wikidata.org/wiki/Q891179","display_name":"Data loss","level":2,"score":0.2799000144004822},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.27379998564720154},{"id":"https://openalex.org/C2984118289","wikidata":"https://www.wikidata.org/wiki/Q29954","display_name":"Power consumption","level":3,"score":0.2549000084400177},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.2531000077724457},{"id":"https://openalex.org/C179145077","wikidata":"https://www.wikidata.org/wiki/Q5154130","display_name":"Communication complexity","level":2,"score":0.25200000405311584}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/aicas64808.2025.11173088","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aicas64808.2025.11173088","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":1,"referenced_works":["https://openalex.org/W4399757647"],"related_works":[],"abstract_inverted_index":{"This":[0,150],"paper":[1],"presents":[2],"a":[3,26,62,70,124,153],"novel":[4],"approach":[5],"to":[6,41,87,146,156,169],"enhance":[7],"GPU":[8,121,143],"cluster":[9],"utilization":[10,122,144],"and":[11,69,95,97,119,160],"significantly":[12,168],"reduce":[13,42,134,161],"Large":[14],"Language":[15],"Model":[16],"(LLM)":[17],"training":[18,171],"time":[19],"by":[20,136],"increasing":[21],"checkpointing":[22,109,158],"frequency.":[23],"By":[24],"introducing":[25],"Compute":[27],"Express":[28],"Link":[29],"(CXL)":[30],"based":[31],"PC-DDR":[32],"or":[33,55],"LPDDR":[34],"as":[35],"an":[36,98],"in-memory":[37],"cache,":[38],"we":[39],"aim":[40],"work":[43,151,162],"loss":[44,54,163],"during":[45,164],"node-level":[46,165],"failures,":[47,166],"which":[48],"occur":[49],"more":[50],"frequently":[51],"than":[52],"power":[53],"other":[56],"issues.":[57],"The":[58,111],"proposed":[59],"method":[60],"utilizes":[61],"CXL":[63,83,93],"DRAM":[64,94],"pool":[65],"for":[66,77,105],"frequent":[67,79],"checkpoints":[68,80],"networked":[71],"storage":[72],"lake":[73],"with":[74],"enterprise":[75],"SSDs":[76],"less":[78],"(one-tenth).":[81],"A":[82],"switch":[84],"is":[85,127],"employed":[86],"facilitate":[88],"peer-to-peer":[89],"(P2P)":[90],"communication":[91],"between":[92],"SSD":[96],"elastic":[99],"memory":[100],"model,":[101],"making":[102],"it":[103,139],"suitable":[104],"various":[106],"types":[107],"of":[108,113],"techniques.":[110],"impact":[112],"model":[114,131],"parallelization":[115,132],"on":[116],"read/write":[117],"bandwidth":[118,135],"efficient":[120],"at":[123],"node":[125],"level":[126],"also":[128,140],"discussed.":[129],"While":[130],"can":[133],"10":[137],"times,":[138],"decreases":[141],"effective":[142],"due":[145],"smaller":[147],"matrix":[148],"computations.":[149],"offers":[152],"practical":[154],"solution":[155],"increase":[157],"frequency":[159],"contributing":[167],"LLM":[170],"efficiency.":[172]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}