{"id":"https://openalex.org/W4401408839","doi":"https://doi.org/10.1145/3673038.3673122","title":"GMM: An Efficient GPU Memory Management-based Model Serving System for Multiple DNN Inference Models","display_name":"GMM: An Efficient GPU Memory Management-based Model Serving System for Multiple DNN Inference Models","publication_year":2024,"publication_date":"2024-08-08","ids":{"openalex":"https://openalex.org/W4401408839","doi":"https://doi.org/10.1145/3673038.3673122"},"language":"en","primary_location":{"id":"doi:10.1145/3673038.3673122","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3673038.3673122","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3673038.3673122","source":null,"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 53rd International Conference on Parallel Processing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3673038.3673122","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5005892736","display_name":"XinYu Piao","orcid":"https://orcid.org/0009-0000-7502-4080"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"XinYu Piao","raw_affiliation_strings":["Korea University, South Korea"],"raw_orcid":"https://orcid.org/0009-0000-7502-4080","affiliations":[{"raw_affiliation_string":"Korea University, South Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5018364461","display_name":"Jong\u2010Kook Kim","orcid":"https://orcid.org/0000-0003-1828-7807"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jong-Kook Kim","raw_affiliation_strings":["Korea University, South Korea"],"raw_orcid":"https://orcid.org/0000-0003-1828-7807","affiliations":[{"raw_affiliation_string":"Korea University, South Korea","institution_ids":["https://openalex.org/I197347611"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I197347611"],"apc_list":null,"apc_paid":null,"fwci":0.2142,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.48429074,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"660","last_page":"668"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9901000261306763,"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":0.9901000261306763,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9781000018119812,"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"}},{"id":"https://openalex.org/T10444","display_name":"Context-Aware Activity Recognition Systems","score":0.9641000032424927,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8010393381118774},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6839959025382996},{"id":"https://openalex.org/keywords/memory-management","display_name":"Memory management","score":0.5444435477256775},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4390207529067993},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3795863389968872},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.36271941661834717},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.1846102476119995},{"id":"https://openalex.org/keywords/semiconductor-memory","display_name":"Semiconductor memory","score":0.14761394262313843}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8010393381118774},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6839959025382996},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.5444435477256775},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4390207529067993},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3795863389968872},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.36271941661834717},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.1846102476119995},{"id":"https://openalex.org/C98986596","wikidata":"https://www.wikidata.org/wiki/Q1143031","display_name":"Semiconductor memory","level":2,"score":0.14761394262313843}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3673038.3673122","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3673038.3673122","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3673038.3673122","source":null,"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 53rd International Conference on Parallel Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3673038.3673122","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3673038.3673122","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3673038.3673122","source":null,"license":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 53rd International Conference on Parallel Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.7400000095367432,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G1571933249","display_name":null,"funder_award_id":"2016R1D1A1B04933156","funder_id":"https://openalex.org/F4320323817","funder_display_name":"Universitas Brawijaya"}],"funders":[{"id":"https://openalex.org/F4320323817","display_name":"Universitas Brawijaya","ror":"https://ror.org/01wk3d929"}],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4401408839.pdf"},"referenced_works_count":10,"referenced_works":["https://openalex.org/W95608104","https://openalex.org/W2097117768","https://openalex.org/W2194775991","https://openalex.org/W2734941459","https://openalex.org/W2970971581","https://openalex.org/W2982083293","https://openalex.org/W4321853806","https://openalex.org/W4372263604","https://openalex.org/W4387321503","https://openalex.org/W6713134421"],"related_works":["https://openalex.org/W2055243143","https://openalex.org/W4321636575","https://openalex.org/W1986418932","https://openalex.org/W2357796999","https://openalex.org/W2045526782","https://openalex.org/W2741131631","https://openalex.org/W2156919374","https://openalex.org/W1984019423","https://openalex.org/W4280588203","https://openalex.org/W1483472507"],"abstract_inverted_index":{"Recent":[0],"DNN":[1],"model":[2,27,88],"serving":[3,28,89],"systems":[4,12,29,69],"have":[5],"begun":[6],"to":[7,13,22,93,114,118,129,155,169],"use":[8],"multi-GPUs":[9],"and":[10,46,149,184],"distributed":[11],"serve":[14,94],"a":[15,63,111,174],"variety":[16],"of":[17],"inference":[18,33,43,60,77,96,132,167,189],"models":[19,34,44,78,97,117,133,168],"as":[20,72,110],"services":[21],"users.":[23],"However,":[24],"modern":[25],"GPU-based":[26],"cannot":[30,55],"execute":[31,47,135],"multiple":[32,95],"beyond":[35,98],"the":[36,73,99,106,122,137,142,151],"GPU":[37,52,85,100,107,123,138,146,176],"memory":[38,53,101,108,124,147,152],"size.":[39],"This":[40,80],"is":[41],"because":[42],"occupy":[45],"on":[48,136,173],"their":[49],"own":[50],"pre-allocated":[51],"that":[54,131],"be":[56,170],"shared":[57],"with":[58],"other":[59],"models.":[61],"As":[62],"result,":[64],"more":[65,76,166],"GPUs":[66],"or":[67],"large":[68,112],"are":[70],"required":[71],"demand":[74],"for":[75,159,165],"increases.":[79],"paper":[81],"proposes":[82],"an":[83],"efficient":[84],"Memory":[86],"Management-based":[87],"system,":[90],"called":[91],"GMM,":[92],"limit.":[102],"The":[103,161],"GMM":[104,143],"initializes":[105],"space":[109,148],"tensor":[113],"allow":[115],"all":[116],"cache":[119,156],"anywhere":[120],"in":[121,181,185],"without":[125,139],"any":[126,140],"constraints.":[127],"Then":[128],"ensure":[130],"can":[134],"conflicts,":[141],"finds":[144],"unused":[145],"uses":[150],"overwriting":[153],"method":[154],"model\u2019s":[157],"parameters":[158],"execution.":[160],"proposed":[162],"system":[163],"allows":[164],"executed":[171],"parallel":[172],"single":[175],"than":[177],"previous":[178],"systems,":[179],"resulting":[180],"higher":[182],"throughput":[183],"some":[186],"cases,":[187],"shorter":[188],"time.":[190]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}