{"id":"https://openalex.org/W4404954309","doi":"https://doi.org/10.1109/micro61859.2024.00021","title":"vTrain: A Simulation Framework for Evaluating Cost-Effective and Compute-Optimal Large Language Model Training","display_name":"vTrain: A Simulation Framework for Evaluating Cost-Effective and Compute-Optimal Large Language Model Training","publication_year":2024,"publication_date":"2024-11-02","ids":{"openalex":"https://openalex.org/W4404954309","doi":"https://doi.org/10.1109/micro61859.2024.00021"},"language":"en","primary_location":{"id":"doi:10.1109/micro61859.2024.00021","is_oa":false,"landing_page_url":"https://doi.org/10.1109/micro61859.2024.00021","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 57th IEEE/ACM International Symposium on Microarchitecture (MICRO)","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/A5113077693","display_name":"J. Bang","orcid":null},"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":"Jehyeon Bang","raw_affiliation_strings":["KAIST, School of Electrical Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST, School of Electrical Engineering","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101073011","display_name":"Yujeong Choi","orcid":null},"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":"Yujeong Choi","raw_affiliation_strings":["KAIST, School of Electrical Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST, School of Electrical Engineering","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058815446","display_name":"Myeongwoo Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Myeongwoo Kim","raw_affiliation_strings":["Samsung Advanced Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Advanced Institute of Technology","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047323054","display_name":"Yongdeok Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I2250650973","display_name":"Samsung (South Korea)","ror":"https://ror.org/04w3jy968","country_code":"KR","type":"company","lineage":["https://openalex.org/I2250650973"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yongdeok Kim","raw_affiliation_strings":["Samsung Advanced Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Advanced Institute of Technology","institution_ids":["https://openalex.org/I2250650973"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5091648103","display_name":"Minsoo Rhu","orcid":"https://orcid.org/0000-0003-3303-8681"},"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":"Minsoo Rhu","raw_affiliation_strings":["KAIST, School of Electrical Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST, School of Electrical Engineering","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":3.0563,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":{"value":0.92611054,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"153","last_page":"167"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.982699990272522,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.982699990272522,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9696999788284302,"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/computer-science","display_name":"Computer science","score":0.7810055613517761},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.6817828416824341},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5118376612663269},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3667184114456177}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7810055613517761},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.6817828416824341},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5118376612663269},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3667184114456177},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/micro61859.2024.00021","is_oa":false,"landing_page_url":"https://doi.org/10.1109/micro61859.2024.00021","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 57th IEEE/ACM International Symposium on Microarchitecture (MICRO)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":62,"referenced_works":["https://openalex.org/W2057332538","https://openalex.org/W2157225945","https://openalex.org/W2798515322","https://openalex.org/W2883830791","https://openalex.org/W2969388332","https://openalex.org/W2981114289","https://openalex.org/W3011751313","https://openalex.org/W3016939927","https://openalex.org/W3022298203","https://openalex.org/W3081168214","https://openalex.org/W3086105743","https://openalex.org/W3092260380","https://openalex.org/W3096425133","https://openalex.org/W3097320606","https://openalex.org/W3129831491","https://openalex.org/W3132107458","https://openalex.org/W3158444059","https://openalex.org/W3197816522","https://openalex.org/W3204998121","https://openalex.org/W3205803342","https://openalex.org/W3206418100","https://openalex.org/W3209503812","https://openalex.org/W4281790033","https://openalex.org/W4290991121","https://openalex.org/W4318541537","https://openalex.org/W4318541676","https://openalex.org/W4318541692","https://openalex.org/W4381886086","https://openalex.org/W4381894573","https://openalex.org/W4387321109","https://openalex.org/W4387951242","https://openalex.org/W4388031315","https://openalex.org/W6739901393","https://openalex.org/W6753584990","https://openalex.org/W6756009870","https://openalex.org/W6756718674","https://openalex.org/W6758283263","https://openalex.org/W6759814162","https://openalex.org/W6767997687","https://openalex.org/W6769424276","https://openalex.org/W6778883912","https://openalex.org/W6779076691","https://openalex.org/W6779863968","https://openalex.org/W6780805062","https://openalex.org/W6781728138","https://openalex.org/W6782839094","https://openalex.org/W6784871562","https://openalex.org/W6784999070","https://openalex.org/W6787953186","https://openalex.org/W6793903029","https://openalex.org/W6810081322","https://openalex.org/W6810220367","https://openalex.org/W6810296985","https://openalex.org/W6810738896","https://openalex.org/W6810743949","https://openalex.org/W6811340617","https://openalex.org/W6811928498","https://openalex.org/W6838322825","https://openalex.org/W6850625674","https://openalex.org/W6854308872","https://openalex.org/W6854866820","https://openalex.org/W6860041859"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W230091440","https://openalex.org/W2390279801","https://openalex.org/W2233261550","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2810751659"],"abstract_inverted_index":{"As":[0],"large":[1,24],"language":[2],"models":[3,26],"(LLMs)":[4],"become":[5],"widespread":[6],"in":[7,27],"various":[8],"application":[9],"domains,":[10],"a":[11,28,37,53,95,147,153],"critical":[12],"challenge":[13],"the":[14,57,74],"AI":[15,25,93],"community":[16],"is":[17,19,44],"facing":[18],"how":[20],"to":[21,68,101],"train":[22],"these":[23],"cost-effective":[29,106],"manner.":[30],"Existing":[31],"LLM":[32,61,107,142,149],"training":[33,41,82,108,123,128,133,143],"plans":[34],"typically":[35],"employ":[36],"heuristic":[38],"based":[39,45],"parallel":[40],"strategy":[42],"which":[43],"on":[46,73],"empirical":[47],"observations":[48],"rather":[49],"than":[50],"grounded":[51],"upon":[52],"thorough":[54],"examination":[55],"of":[56,60,78,81],"search":[58],"space":[59],"parallelization.":[62],"Such":[63],"limitation":[64],"renders":[65],"existing":[66],"systems":[67],"leave":[69],"significant":[70],"performance":[71],"left":[72],"table,":[75],"wasting":[76],"millions":[77],"dollars":[79],"worth":[80],"cost.":[83],"This":[84],"paper":[85],"presents":[86],"our":[87],"profiling-driven":[88],"simulator":[89],"called":[90],"vTrain,":[91],"providing":[92],"practitioners":[94],"fast":[96],"yet":[97],"accurate":[98],"software":[99],"framework":[100],"determine":[102],"an":[103],"efficient":[104,135],"and":[105,130,145],"system":[109],"configuration.":[110],"We":[111],"demonstrate":[112],"vTrain's":[113],"practicality":[114],"through":[115],"several":[116],"case":[117],"studies,":[118],"e.g.,":[119],"effectively":[120],"evaluating":[121],"optimal":[122],"parallelization":[124],"strategies":[125],"that":[126],"balances":[127],"time":[129],"its":[131],"associated":[132],"cost,":[134],"multi-tenant":[136],"GPU":[137],"cluster":[138],"schedulers":[139],"targeting":[140],"multiple":[141],"jobs,":[144],"determining":[146],"compute-optimal":[148],"model":[150],"architecture":[151],"given":[152],"fixed":[154],"compute":[155],"budget.":[156]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":7}],"updated_date":"2025-12-19T19:40:27.379048","created_date":"2025-10-10T00:00:00"}