{"id":"https://openalex.org/W2798707674","doi":"https://doi.org/10.23919/date.2018.8341972","title":"moDNN: Memory optimal DNN training on GPUs","display_name":"moDNN: Memory optimal DNN training on GPUs","publication_year":2018,"publication_date":"2018-03-01","ids":{"openalex":"https://openalex.org/W2798707674","doi":"https://doi.org/10.23919/date.2018.8341972","mag":"2798707674"},"language":"en","primary_location":{"id":"doi:10.23919/date.2018.8341972","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date.2018.8341972","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 Design, Automation & Test in Europe Conference & Exhibition (DATE)","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/A5100373745","display_name":"Xiaohong Chen","orcid":"https://orcid.org/0000-0002-9797-8384"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["CN","US"],"is_corresponding":true,"raw_author_name":"Xiaoming Chen","raw_affiliation_strings":["Department of Computer Science and Engineering, University of Notre Dame, IN, USA","State Key Laboratory of Computer Architecture, Chinese Academy of Sciences, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University of Notre Dame, IN, USA","institution_ids":["https://openalex.org/I107639228"]},{"raw_affiliation_string":"State Key Laboratory of Computer Architecture, Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060901632","display_name":"Danny Z. Chen","orcid":"https://orcid.org/0000-0001-6565-2884"},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Danny Z. Chen","raw_affiliation_strings":["Department of Computer Science and Engineering, University of Notre Dame, IN, USA"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University of Notre Dame, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100600905","display_name":"Xiaobo Sharon Hu","orcid":"https://orcid.org/0000-0002-6636-9738"},"institutions":[{"id":"https://openalex.org/I107639228","display_name":"University of Notre Dame","ror":"https://ror.org/00mkhxb43","country_code":"US","type":"education","lineage":["https://openalex.org/I107639228"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiaobo Sharon Hu","raw_affiliation_strings":["Department of Computer Science and Engineering, University of Notre Dame, IN, USA"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, University of Notre Dame, IN, USA","institution_ids":["https://openalex.org/I107639228"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5100373745"],"corresponding_institution_ids":["https://openalex.org/I107639228","https://openalex.org/I19820366"],"apc_list":null,"apc_paid":null,"fwci":1.6713,"has_fulltext":false,"cited_by_count":26,"citation_normalized_percentile":{"value":0.88331446,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"13","last_page":"18"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9941999912261963,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9909999966621399,"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":0.9001362323760986},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6771167516708374},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6722743511199951},{"id":"https://openalex.org/keywords/schedule","display_name":"Schedule","score":0.5140630006790161},{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.47963815927505493},{"id":"https://openalex.org/keywords/graphics","display_name":"Graphics","score":0.4693390429019928},{"id":"https://openalex.org/keywords/memory-management","display_name":"Memory management","score":0.4363780915737152},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.42603054642677307},{"id":"https://openalex.org/keywords/high-memory","display_name":"High memory","score":0.4169573187828064},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.41382527351379395},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.32171154022216797},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.21939119696617126},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.18907985091209412},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.17831632494926453},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.12757351994514465},{"id":"https://openalex.org/keywords/semiconductor-memory","display_name":"Semiconductor memory","score":0.09861773252487183}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.9001362323760986},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6771167516708374},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6722743511199951},{"id":"https://openalex.org/C68387754","wikidata":"https://www.wikidata.org/wiki/Q7271585","display_name":"Schedule","level":2,"score":0.5140630006790161},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.47963815927505493},{"id":"https://openalex.org/C21442007","wikidata":"https://www.wikidata.org/wiki/Q1027879","display_name":"Graphics","level":2,"score":0.4693390429019928},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.4363780915737152},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.42603054642677307},{"id":"https://openalex.org/C2781357197","wikidata":"https://www.wikidata.org/wiki/Q5757597","display_name":"High memory","level":2,"score":0.4169573187828064},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.41382527351379395},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.32171154022216797},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.21939119696617126},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.18907985091209412},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.17831632494926453},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.12757351994514465},{"id":"https://openalex.org/C98986596","wikidata":"https://www.wikidata.org/wiki/Q1143031","display_name":"Semiconductor memory","level":2,"score":0.09861773252487183}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.23919/date.2018.8341972","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date.2018.8341972","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2018 Design, Automation & Test in Europe Conference & Exhibition (DATE)","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":38,"referenced_works":["https://openalex.org/W128208790","https://openalex.org/W1667652561","https://openalex.org/W1686810756","https://openalex.org/W1841592590","https://openalex.org/W1903029394","https://openalex.org/W1922123711","https://openalex.org/W1972501001","https://openalex.org/W2046441184","https://openalex.org/W2097117768","https://openalex.org/W2102017903","https://openalex.org/W2102605133","https://openalex.org/W2163605009","https://openalex.org/W2172654076","https://openalex.org/W2194775991","https://openalex.org/W2286365479","https://openalex.org/W2300242332","https://openalex.org/W2319920447","https://openalex.org/W2332506150","https://openalex.org/W2489529491","https://openalex.org/W2884834742","https://openalex.org/W2962776038","https://openalex.org/W2963042536","https://openalex.org/W2963340555","https://openalex.org/W2963374099","https://openalex.org/W2963674932","https://openalex.org/W2964174152","https://openalex.org/W3152703111","https://openalex.org/W4293404878","https://openalex.org/W4297827580","https://openalex.org/W4302296459","https://openalex.org/W6638632666","https://openalex.org/W6638783484","https://openalex.org/W6675112735","https://openalex.org/W6684191040","https://openalex.org/W6696004547","https://openalex.org/W6698200048","https://openalex.org/W6717768942","https://openalex.org/W7057659579"],"related_works":["https://openalex.org/W2141978748","https://openalex.org/W2612219836","https://openalex.org/W4200309432","https://openalex.org/W1981291981","https://openalex.org/W4253567283","https://openalex.org/W3092905871","https://openalex.org/W2567721291","https://openalex.org/W2896122000","https://openalex.org/W1527643973","https://openalex.org/W2080461002"],"abstract_inverted_index":{"Graphics":[0],"processing":[1],"units":[2],"(GPUs)":[3],"are":[4],"widely":[5],"adopted":[6],"to":[7,44,60,87,99,127,143,180],"accelerate":[8],"the":[9,17,23,46,69,101,117,123,132,138,156,170],"training":[10,58,96],"of":[11,26,56,77],"deep":[12],"neural":[13],"networks":[14],"(DNNs).":[15],"However,":[16],"limited":[18],"GPU":[19,139,152],"memory":[20,47,64,102,118,124,140],"size":[21,110],"restricts":[22],"maximum":[24],"scale":[25],"DNNs":[27],"that":[28,137,167],"can":[29,121],"be":[30],"trained":[31],"on":[32,150,189],"GPUs,":[33],"which":[34,113,135,162],"presents":[35],"serious":[36],"challenges.":[37],"This":[38],"paper":[39],"proposes":[40],"an":[41],"moDNN":[42,52,120,149,176],"framework":[43],"optimize":[45,100],"usage":[48,125],"in":[49],"DNN":[50,57],"training.":[51],"supports":[53],"automatic":[54],"tuning":[55],"code":[59],"match":[61],"any":[62],"given":[63],"budget":[65],"(not":[66],"smaller":[67],"than":[68,166],"theoretical":[70],"lower":[71,165],"bound).":[72],"By":[73],"taking":[74],"full":[75],"advantage":[76],"overlapping":[78],"computations":[79],"and":[80,92,183],"data":[81,90],"transfers,":[82],"we":[83],"have":[84],"developed":[85],"heuristics":[86],"judiciously":[88],"schedule":[89],"offloading":[91],"prefetching,":[93],"together":[94],"with":[95,131,153],"algorithm":[97],"selection,":[98],"usage.":[103,119],"We":[104],"further":[105],"introduce":[106],"a":[107,151],"new":[108],"sub-batch":[109],"selection":[111],"method":[112],"also":[114,178],"greatly":[115],"reduces":[116],"save":[122],"up":[126],"50":[128],"\u03c7,":[129],"compared":[130],"ideal":[133],"case":[134],"assumes":[136],"is":[141,159,163,177],"sufficient":[142],"hold":[144],"all":[145],"data.":[146],"When":[147],"executing":[148],"12GB":[154],"memory,":[155],"performance":[157],"loss":[158],"only":[160],"8%,":[161],"much":[164],"caused":[168],"by":[169],"best":[171],"known":[172],"existing":[173],"approach,":[174],"vDNN.":[175],"applicable":[179],"multiple":[181],"GPUs":[182],"attains":[184],"1.84":[185],"\u03c7":[186],"average":[187],"speedup":[188],"two":[190],"GPUs.":[191]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":7},{"year":2020,"cited_by_count":4},{"year":2019,"cited_by_count":4},{"year":2018,"cited_by_count":1}],"updated_date":"2026-01-17T23:10:49.606395","created_date":"2025-10-10T00:00:00"}