{"id":"https://openalex.org/W2954698599","doi":"https://doi.org/10.1145/3315573.3329984","title":"Automatic GPU memory management for large neural models in TensorFlow","display_name":"Automatic GPU memory management for large neural models in TensorFlow","publication_year":2019,"publication_date":"2019-06-07","ids":{"openalex":"https://openalex.org/W2954698599","doi":"https://doi.org/10.1145/3315573.3329984","mag":"2954698599"},"language":"en","primary_location":{"id":"doi:10.1145/3315573.3329984","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3315573.3329984","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2019 ACM SIGPLAN International Symposium on Memory Management","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/A5089550246","display_name":"Tung D. Le","orcid":null},"institutions":[{"id":"https://openalex.org/I4210145865","display_name":"IBM Research - Tokyo","ror":"https://ror.org/04915qk43","country_code":"JP","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115","https://openalex.org/I4210145865"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Tung D. Le","raw_affiliation_strings":["IBM Research, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Japan","institution_ids":["https://openalex.org/I4210145865"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102286597","display_name":"Haruki Imai","orcid":null},"institutions":[{"id":"https://openalex.org/I4210145865","display_name":"IBM Research - Tokyo","ror":"https://ror.org/04915qk43","country_code":"JP","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115","https://openalex.org/I4210145865"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Haruki Imai","raw_affiliation_strings":["IBM Research, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Japan","institution_ids":["https://openalex.org/I4210145865"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090981513","display_name":"Yasushi Negishi","orcid":null},"institutions":[{"id":"https://openalex.org/I4210145865","display_name":"IBM Research - Tokyo","ror":"https://ror.org/04915qk43","country_code":"JP","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115","https://openalex.org/I4210145865"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Yasushi Negishi","raw_affiliation_strings":["IBM Research, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Japan","institution_ids":["https://openalex.org/I4210145865"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5063074861","display_name":"Kiyokuni Kawachiya","orcid":null},"institutions":[{"id":"https://openalex.org/I4210145865","display_name":"IBM Research - Tokyo","ror":"https://ror.org/04915qk43","country_code":"JP","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210114115","https://openalex.org/I4210145865"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Kiyokuni Kawachiya","raw_affiliation_strings":["IBM Research, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research, Japan","institution_ids":["https://openalex.org/I4210145865"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.0169,"has_fulltext":false,"cited_by_count":17,"citation_normalized_percentile":{"value":0.80748579,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"13"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9991000294685364,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9991000294685364,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T12292","display_name":"Graph Theory and Algorithms","score":0.9962999820709229,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.9929999709129333,"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":0.8557194471359253},{"id":"https://openalex.org/keywords/rewriting","display_name":"Rewriting","score":0.7403922080993652},{"id":"https://openalex.org/keywords/swap","display_name":"Swap (finance)","score":0.7046315670013428},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5794689655303955},{"id":"https://openalex.org/keywords/graph-rewriting","display_name":"Graph rewriting","score":0.4727514684200287},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.46225661039352417},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4518801271915436},{"id":"https://openalex.org/keywords/topological-sorting","display_name":"Topological sorting","score":0.4476704001426697},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.43961814045906067},{"id":"https://openalex.org/keywords/memory-model","display_name":"Memory model","score":0.4351629912853241},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.37592750787734985},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.2761904001235962},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.22326406836509705},{"id":"https://openalex.org/keywords/directed-acyclic-graph","display_name":"Directed acyclic graph","score":0.17884328961372375},{"id":"https://openalex.org/keywords/shared-memory","display_name":"Shared memory","score":0.1644614040851593}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8557194471359253},{"id":"https://openalex.org/C154690210","wikidata":"https://www.wikidata.org/wiki/Q1668499","display_name":"Rewriting","level":2,"score":0.7403922080993652},{"id":"https://openalex.org/C99821215","wikidata":"https://www.wikidata.org/wiki/Q1136583","display_name":"Swap (finance)","level":2,"score":0.7046315670013428},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5794689655303955},{"id":"https://openalex.org/C558772884","wikidata":"https://www.wikidata.org/wiki/Q1508564","display_name":"Graph rewriting","level":3,"score":0.4727514684200287},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.46225661039352417},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4518801271915436},{"id":"https://openalex.org/C176032523","wikidata":"https://www.wikidata.org/wiki/Q753127","display_name":"Topological sorting","level":3,"score":0.4476704001426697},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.43961814045906067},{"id":"https://openalex.org/C12186640","wikidata":"https://www.wikidata.org/wiki/Q6815743","display_name":"Memory model","level":3,"score":0.4351629912853241},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37592750787734985},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2761904001235962},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.22326406836509705},{"id":"https://openalex.org/C74197172","wikidata":"https://www.wikidata.org/wiki/Q1195339","display_name":"Directed acyclic graph","level":2,"score":0.17884328961372375},{"id":"https://openalex.org/C133875982","wikidata":"https://www.wikidata.org/wiki/Q764810","display_name":"Shared memory","level":2,"score":0.1644614040851593},{"id":"https://openalex.org/C10138342","wikidata":"https://www.wikidata.org/wiki/Q43015","display_name":"Finance","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3315573.3329984","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3315573.3329984","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2019 ACM SIGPLAN International Symposium on Memory Management","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":13,"referenced_works":["https://openalex.org/W2186615578","https://openalex.org/W2302255633","https://openalex.org/W2338908902","https://openalex.org/W2489529491","https://openalex.org/W2552911551","https://openalex.org/W2618530766","https://openalex.org/W2621550233","https://openalex.org/W2783292547","https://openalex.org/W2785745756","https://openalex.org/W2788491273","https://openalex.org/W2949650786","https://openalex.org/W2964132039","https://openalex.org/W4301361180"],"related_works":["https://openalex.org/W2009172267","https://openalex.org/W60870644","https://openalex.org/W2140997306","https://openalex.org/W2542922519","https://openalex.org/W4281720143","https://openalex.org/W2619273670","https://openalex.org/W1590684856","https://openalex.org/W3128019261","https://openalex.org/W3005022890","https://openalex.org/W2954698599"],"abstract_inverted_index":{"Deep":[0],"learning":[1],"models":[2],"are":[3,32,84],"becoming":[4],"larger":[5,204,213],"and":[6,37,41,81,157,164,207],"will":[7],"not":[8],"fit":[9],"in":[10,35,53,118,144,191],"the":[11,51,73,104,119,124,138,169,184],"limited":[12],"memory":[13,105],"of":[14,70,76,107,126,140],"accelerators":[15],"such":[16],"as":[17],"GPUs":[18],"for":[19,62,100,183],"training.":[20],"Though":[21],"many":[22],"methods":[23],"have":[24],"been":[25],"proposed":[26],"to":[27,39,57,86,137,153,167,177],"solve":[28],"this":[29,47],"problem,":[30],"they":[31],"rather":[33],"ad-hoc":[34],"nature":[36],"difficult":[38],"extend":[40],"integrate":[42],"with":[43,201,210],"other":[44],"techniques.":[45],"In":[46],"paper,":[48],"we":[49,172,197],"tackle":[50],"problem":[52,125,139],"a":[54,59,68,77,96,108,122,128,132,145,174,189,202,211],"formal":[55],"way":[56],"provide":[58],"strong":[60],"foundation":[61],"supporting":[63],"large":[64,129],"models.":[65],"We":[66,149,187],"propose":[67,173],"method":[69],"formally":[71,154],"rewriting":[72],"computational":[74],"graph":[75,102,163],"model":[78,109,130],"where":[79],"swap-out":[80,156],"swap-in":[82,158],"operations":[83,159],"inserted":[85],"temporarily":[87],"store":[88],"intermediate":[89],"results":[90],"on":[91],"CPU":[92],"memory.":[93],"By":[94],"introducing":[95],"categorized":[97,146],"topological":[98,147],"ordering":[99],"simulating":[101],"execution,":[103],"consumption":[106],"can":[110],"be":[111],"easily":[112],"analyzed":[113],"by":[114,195],"using":[115],"operation":[116,142],"distances":[117,143],"ordering.":[120,148],"As":[121],"result,":[123],"fitting":[127],"into":[131],"memory-limited":[133],"accelerator":[134],"is":[135],"reduced":[136],"reducing":[141],"then":[150],"show":[151],"how":[152],"derive":[155],"from":[160],"an":[161],"existing":[162],"present":[165],"rules":[166],"optimize":[168],"graph.":[170],"Finally,":[171],"simulation-based":[175],"auto-tuning":[176],"automatically":[178],"find":[179],"suitable":[180],"graph-rewriting":[181],"parameters":[182],"best":[185],"performance.":[186],"developed":[188],"module":[190],"TensorFlow,":[192],"called":[193],"LMS,":[194],"which":[196],"successfully":[198],"trained":[199],"ResNet-50":[200],"4.9x":[203],"mini-batch":[205],"size":[206],"3D":[208],"U-Net":[209],"5.6x":[212],"image":[214],"resolution.":[215]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":2},{"year":2020,"cited_by_count":4},{"year":2019,"cited_by_count":2}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}