{"id":"https://openalex.org/W7116442584","doi":"https://doi.org/10.1145/3754598.3754605","title":"MixLoRA: An Efficient Multi-Tenant Framework for Concurrently Serving Diverse LoRA Models in Large Language Models","display_name":"MixLoRA: An Efficient Multi-Tenant Framework for Concurrently Serving Diverse LoRA Models in Large Language Models","publication_year":2025,"publication_date":"2025-09-08","ids":{"openalex":"https://openalex.org/W7116442584","doi":"https://doi.org/10.1145/3754598.3754605"},"language":null,"primary_location":{"id":"doi:10.1145/3754598.3754605","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3754598.3754605","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th 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://doi.org/10.1145/3754598.3754605","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5056476277","display_name":"Ronghuai Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I162868743","display_name":"Tianjin University","ror":"https://ror.org/012tb2g32","country_code":"CN","type":"education","lineage":["https://openalex.org/I162868743"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Ronghuai Chen","raw_affiliation_strings":["Tianjin University, Tianjin, China"],"affiliations":[{"raw_affiliation_string":"Tianjin University, Tianjin, China","institution_ids":["https://openalex.org/I162868743"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120907155","display_name":"Ce Yu","orcid":null},"institutions":[{"id":"https://openalex.org/I162868743","display_name":"Tianjin University","ror":"https://ror.org/012tb2g32","country_code":"CN","type":"education","lineage":["https://openalex.org/I162868743"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ce Yu","raw_affiliation_strings":["Tianjin University, Tianjin, China"],"affiliations":[{"raw_affiliation_string":"Tianjin University, Tianjin, China","institution_ids":["https://openalex.org/I162868743"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115501032","display_name":"Hao Fu","orcid":"https://orcid.org/0000-0002-4349-8748"},"institutions":[{"id":"https://openalex.org/I4210165734","display_name":"National Supercomputing Center of Tianjin","ror":"https://ror.org/05tngxm14","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210165734"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Fu","raw_affiliation_strings":["National Supercomputing Center in Tianjin, Tianjin, China"],"affiliations":[{"raw_affiliation_string":"National Supercomputing Center in Tianjin, Tianjin, China","institution_ids":["https://openalex.org/I4210165734"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112426356","display_name":"Xiaoteng Hu","orcid":null},"institutions":[{"id":"https://openalex.org/I162868743","display_name":"Tianjin University","ror":"https://ror.org/012tb2g32","country_code":"CN","type":"education","lineage":["https://openalex.org/I162868743"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaoteng Hu","raw_affiliation_strings":["Tianjin University, Tianjin, China"],"affiliations":[{"raw_affiliation_string":"Tianjin University, Tianjin, China","institution_ids":["https://openalex.org/I162868743"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5120894016","display_name":"Bin Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I162868743","display_name":"Tianjin University","ror":"https://ror.org/012tb2g32","country_code":"CN","type":"education","lineage":["https://openalex.org/I162868743"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bin Yang","raw_affiliation_strings":["Tianjin University, Tianjin, China"],"affiliations":[{"raw_affiliation_string":"Tianjin University, Tianjin, China","institution_ids":["https://openalex.org/I162868743"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5056476277"],"corresponding_institution_ids":["https://openalex.org/I162868743"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.63371236,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11","last_page":"21"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.1941000074148178,"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":0.1941000074148178,"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/T14347","display_name":"Big Data and Digital Economy","score":0.1168999969959259,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.06960000097751617,"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/scalability","display_name":"Scalability","score":0.7623000144958496},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.7113000154495239},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.7103000283241272},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.689300000667572},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.5322999954223633},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.41359999775886536},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.3944999873638153}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8672000169754028},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7623000144958496},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.7113000154495239},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.7103000283241272},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.689300000667572},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.5534999966621399},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.5322999954223633},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.41359999775886536},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.3944999873638153},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.34689998626708984},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.3416999876499176},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.32600000500679016},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.3075000047683716},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.29980000853538513},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.2924000024795532},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.2921000123023987},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.2678999900817871},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.2621000111103058},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.259799987077713}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3754598.3754605","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3754598.3754605","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th International Conference on Parallel Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3754598.3754605","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3754598.3754605","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th International Conference on Parallel Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.4405056834220886,"display_name":"Decent work and economic growth","id":"https://metadata.un.org/sdg/8"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":6,"referenced_works":["https://openalex.org/W3168867926","https://openalex.org/W4322718191","https://openalex.org/W4379539933","https://openalex.org/W4385245566","https://openalex.org/W4387321091","https://openalex.org/W4415797413"],"related_works":[],"abstract_inverted_index":{"The":[0],"rapid":[1],"advancement":[2],"of":[3,13,68,83,103],"large":[4],"language":[5],"models":[6],"(LLMs)":[7],"has":[8],"driven":[9],"the":[10,80,90,108],"widespread":[11],"adoption":[12],"Low-Rank":[14],"Adaptation":[15],"(LoRA)":[16],"for":[17,85,110,142],"efficient":[18,140],"fine-tuning.":[19],"However,":[20],"existing":[21],"multi-LoRA":[22],"inference":[23,93,145],"systems,":[24],"such":[25],"as":[26],"Punica,":[27],"face":[28],"significant":[29],"challenges":[30],"in":[31],"GPU":[32,62],"utilization":[33,116],"when":[34],"handling":[35],"concurrent":[36,86],"requests":[37,102],"with":[38],"diverse":[39],"ranks,":[40],"leading":[41],"to":[42,128,133],"suboptimal":[43],"throughput":[44,131],"and":[45,96,113,139],"increased":[46],"latency.":[47],"To":[48],"address":[49],"these":[50],"limitations,":[51],"we":[52],"propose":[53],"MixLoRA,":[54],"a":[55,74,104,137],"novel":[56],"CUDA":[57,119],"stream-based":[58],"framework":[59],"that":[60,78,124],"enhances":[61],"efficiency":[63],"by":[64],"enabling":[65],"parallel":[66],"execution":[67,87],"mixed-rank":[69],"LoRA":[70],"requests.":[71],"MixLoRA":[72,125],"introduces":[73],"multi-worker":[75],"scheduling":[76],"architecture":[77],"selects":[79],"optimal":[81],"number":[82],"workers":[84],"based":[88],"on":[89],"model":[91],"size,":[92],"request":[94],"characteristics,":[95],"device":[97],"information.":[98],"Each":[99],"worker":[100],"manages":[101],"fixed":[105],"rank,":[106],"eliminating":[107],"need":[109],"rank-aligned":[111],"batching":[112],"maximizing":[114],"resource":[115],"through":[117],"independent":[118],"streams.":[120],"Experimental":[121],"results":[122],"demonstrate":[123],"achieves":[126],"up":[127],"1.2\u00d7":[129],"higher":[130],"compared":[132],"state-of-the-art":[134],"solutions,":[135],"offering":[136],"scalable":[138],"approach":[141],"multi-tenant":[143],"LLM":[144],"services.":[146]},"counts_by_year":[],"updated_date":"2025-12-21T02:06:08.432651","created_date":"2025-12-21T00:00:00"}