{"id":"https://openalex.org/W3035078899","doi":"https://doi.org/10.1145/3368089.3417050","title":"Estimating GPU memory consumption of deep learning models","display_name":"Estimating GPU memory consumption of deep learning models","publication_year":2020,"publication_date":"2020-11-08","ids":{"openalex":"https://openalex.org/W3035078899","doi":"https://doi.org/10.1145/3368089.3417050","mag":"3035078899"},"language":"en","primary_location":{"id":"doi:10.1145/3368089.3417050","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3368089.3417050","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering","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/A5031774552","display_name":"Yanjie Gao","orcid":"https://orcid.org/0000-0003-1899-8561"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanjie Gao","raw_affiliation_strings":["Microsoft Research, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004895426","display_name":"Yu Liu","orcid":"https://orcid.org/0000-0002-2067-9175"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yu Liu","raw_affiliation_strings":["Microsoft Research, China / National University of Singapore, Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research, China / National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100412598","display_name":"Hongyu Zhang","orcid":"https://orcid.org/0000-0002-3063-9425"},"institutions":[{"id":"https://openalex.org/I78757542","display_name":"University of Newcastle Australia","ror":"https://ror.org/00eae9z71","country_code":"AU","type":"education","lineage":["https://openalex.org/I78757542"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Hongyu Zhang","raw_affiliation_strings":["University of Newcastle, Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Newcastle, Australia","institution_ids":["https://openalex.org/I78757542"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103288789","display_name":"Zhengxian Li","orcid":null},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhengxian Li","raw_affiliation_strings":["Microsoft Research, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033693404","display_name":"Yonghao Zhu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yonghao Zhu","raw_affiliation_strings":["Microsoft Research, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013547786","display_name":"Haoxiang Lin","orcid":"https://orcid.org/0000-0002-9148-5861"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haoxiang Lin","raw_affiliation_strings":["Microsoft Research, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100438310","display_name":"Mao Yang","orcid":"https://orcid.org/0009-0009-6455-3898"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mao Yang","raw_affiliation_strings":["Microsoft Research, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research, China","institution_ids":["https://openalex.org/I4210113369"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":7.4396,"has_fulltext":false,"cited_by_count":128,"citation_normalized_percentile":{"value":0.97948039,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1342","last_page":"1352"},"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9975000023841858,"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"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9973000288009644,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8594641089439392},{"id":"https://openalex.org/keywords/hyperparameter","display_name":"Hyperparameter","score":0.6925107836723328},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5505139231681824},{"id":"https://openalex.org/keywords/general-purpose-computing-on-graphics-processing-units","display_name":"General-purpose computing on graphics processing units","score":0.48164787888526917},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.46728453040122986},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4519018530845642},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4477463364601135},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.42763829231262207},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.41422903537750244},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.15855249762535095},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.09974503517150879},{"id":"https://openalex.org/keywords/software-engineering","display_name":"Software engineering","score":0.09678515791893005}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8594641089439392},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.6925107836723328},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5505139231681824},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.48164787888526917},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46728453040122986},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4519018530845642},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4477463364601135},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.42763829231262207},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41422903537750244},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.15855249762535095},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.09974503517150879},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.09678515791893005},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3368089.3417050","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3368089.3417050","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering","raw_type":"proceedings-article"},{"id":"pmh:uon:44305","is_oa":false,"landing_page_url":"http://hdl.handle.net/1959.13/1451891","pdf_url":null,"source":{"id":"https://openalex.org/S4377196471","display_name":"NOVA (University of Newcastle Australia)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I78757542","host_organization_name":"University of Newcastle Australia","host_organization_lineage":["https://openalex.org/I78757542"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"conference paper"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth","score":0.6499999761581421}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":38,"referenced_works":["https://openalex.org/W1528966910","https://openalex.org/W1686810756","https://openalex.org/W1983134029","https://openalex.org/W2034996255","https://openalex.org/W2064675550","https://openalex.org/W2072617662","https://openalex.org/W2078490331","https://openalex.org/W2108598243","https://openalex.org/W2143637886","https://openalex.org/W2150355110","https://openalex.org/W2159910937","https://openalex.org/W2183341477","https://openalex.org/W2186615578","https://openalex.org/W2194775991","https://openalex.org/W2338908902","https://openalex.org/W2402144811","https://openalex.org/W2419597278","https://openalex.org/W2425506752","https://openalex.org/W2519870445","https://openalex.org/W2771362265","https://openalex.org/W2787998955","https://openalex.org/W2804032941","https://openalex.org/W2943950766","https://openalex.org/W2953487403","https://openalex.org/W2954141573","https://openalex.org/W2963310665","https://openalex.org/W2963341956","https://openalex.org/W2964121744","https://openalex.org/W2964174152","https://openalex.org/W2967996334","https://openalex.org/W2968594320","https://openalex.org/W2970971581","https://openalex.org/W3005780259","https://openalex.org/W3013672048","https://openalex.org/W3103894541","https://openalex.org/W4240064337","https://openalex.org/W4301361180","https://openalex.org/W4386303051"],"related_works":["https://openalex.org/W2140186469","https://openalex.org/W4280563792","https://openalex.org/W4318719684","https://openalex.org/W4318559728","https://openalex.org/W3183136280","https://openalex.org/W2775233965","https://openalex.org/W4311551265","https://openalex.org/W1980160788","https://openalex.org/W4231288671","https://openalex.org/W2034135205"],"abstract_inverted_index":{"Deep":[0],"learning":[1],"(DL)":[2],"has":[3,74],"been":[4],"increasingly":[5],"adopted":[6],"by":[7,32],"a":[8,33,58,92,99],"variety":[9],"of":[10,24,50,63,86,95,119,134],"software-intensive":[11],"systems.":[12],"Developers":[13],"mainly":[14],"use":[15],"GPUs":[16],"to":[17,39,60,83,91,128],"accelerate":[18],"the":[19,28,42,64,84,131,136,140],"training,":[20],"testing,":[21],"and":[22,68,98,139,162],"deployment":[23],"DL":[25,34,43,78,120,141],"models.":[26,121],"However,":[27],"GPU":[29,66,87,116,174],"memory":[30,67,117,132,175],"consumed":[31],"model":[35],"is":[36,170],"often":[37],"unknown":[38],"them":[40],"before":[41],"job":[44,59,79],"executes.":[45],"Therefore,":[46],"an":[47,111,124],"improper":[48],"choice":[49],"neural":[51],"architecture":[52],"or":[53],"hyperparameters":[54,155],"can":[55],"cause":[56],"such":[57],"run":[61],"out":[62],"limited":[65],"fail.":[69],"Our":[70,164],"recent":[71],"empirical":[72],"study":[73],"found":[75],"that":[76,168],"many":[77],"failures":[80],"are":[81],"due":[82],"exhaustion":[85],"memory.":[88],"This":[89],"leads":[90],"horrendous":[93],"waste":[94],"computing":[96],"resources":[97],"significant":[100],"reduction":[101],"in":[102,172],"development":[103],"productivity.":[104],"In":[105],"this":[106],"paper,":[107],"we":[108],"propose":[109],"DNNMem,":[110],"accurate":[112],"estimation":[113,126],"tool":[114],"for":[115],"consumption":[118,133],"DNNMem":[122,147,169],"employs":[123],"analytic":[125],"approach":[127],"systematically":[129],"calculate":[130],"both":[135],"computation":[137],"graph":[138],"framework":[142],"runtime.":[143],"We":[144],"have":[145],"evaluated":[146],"on":[148],"5":[149],"real-world":[150],"representative":[151],"models":[152],"with":[153],"different":[154],"under":[156],"3":[157],"mainstream":[158],"frameworks":[159],"(TensorFlow,":[160],"PyTorch,":[161],"MXNet).":[163],"extensive":[165],"experiments":[166],"show":[167],"effective":[171],"estimating":[173],"consumption.":[176]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":24},{"year":2024,"cited_by_count":25},{"year":2023,"cited_by_count":27},{"year":2022,"cited_by_count":29},{"year":2021,"cited_by_count":16},{"year":2020,"cited_by_count":4}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}