{"id":"https://openalex.org/W7133536271","doi":"https://doi.org/10.1109/hpca68181.2026.11408552","title":"VectorLiteRAG: Latency-Aware and Fine-Grained Resource Partitioning for Efficient RAG","display_name":"VectorLiteRAG: Latency-Aware and Fine-Grained Resource Partitioning for Efficient RAG","publication_year":2026,"publication_date":"2026-01-31","ids":{"openalex":"https://openalex.org/W7133536271","doi":"https://doi.org/10.1109/hpca68181.2026.11408552"},"language":null,"primary_location":{"id":"doi:10.1109/hpca68181.2026.11408552","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408552","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5122599301","display_name":"Junkyum Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Junkyum Kim","raw_affiliation_strings":["Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089590312","display_name":"Divya Mahajan","orcid":"https://orcid.org/0009-0007-8184-0528"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Divya Mahajan","raw_affiliation_strings":["Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I130701444"],"apc_list":null,"apc_paid":null,"fwci":21.6816,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.97322166,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"15"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.18000000715255737,"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":0.18000000715255737,"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":0.09709999710321426,"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/T10719","display_name":"3D Shape Modeling and Analysis","score":0.03420000150799751,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/resource","display_name":"Resource (disambiguation)","score":0.48030000925064087},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.2694999873638153},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.2635999917984009},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.25519999861717224},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.24480000138282776}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.597100019454956},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.48030000925064087},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2635999917984009},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.25519999861717224},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.24480000138282776},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2442999929189682},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.24240000545978546},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2395000010728836},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.2362000048160553}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpca68181.2026.11408552","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408552","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W2124509324","https://openalex.org/W2138662031","https://openalex.org/W2204555070","https://openalex.org/W2963469388","https://openalex.org/W2970641574","https://openalex.org/W2998702515","https://openalex.org/W3093655911","https://openalex.org/W3174809957","https://openalex.org/W3197720002","https://openalex.org/W3200211247","https://openalex.org/W4280568654","https://openalex.org/W4281634017","https://openalex.org/W4283806888","https://openalex.org/W4387321091","https://openalex.org/W4388778348","https://openalex.org/W4401211704","https://openalex.org/W4401211718","https://openalex.org/W4401857375","https://openalex.org/W4407197060","https://openalex.org/W4410583873","https://openalex.org/W4411471721","https://openalex.org/W4411486516","https://openalex.org/W4414371540","https://openalex.org/W4415124086"],"related_works":[],"abstract_inverted_index":{"Retrieval-Augmented":[0],"Generation":[1],"leverages":[2],"vector":[3,22,50],"similarity":[4],"search":[5,23,51,136],"to":[6,41,95,109,176,179],"enhance":[7],"large":[8],"language":[9],"models":[10],"with":[11,62,122],"up-to-date,":[12],"external":[13],"knowledge,":[14],"enabling":[15,125],"accurate":[16],"and":[17,34,54,71,89,93,106,117,130,138,151,156,186,222],"reliable":[18],"responses.":[19],"While":[20],"CPU-only":[21],"incurs":[24],"high":[25],"latency":[26,137,213],"on":[27,37],"large,":[28],"high-dimensional":[29],"indices,":[30],"co-locating":[31],"the":[32,35,38,120,199,211],"retriever":[33],"LLM":[36,63,123],"GPU":[39,152],"leads":[40],"resource":[42,47],"sharing":[43],"that":[44,84,153,195],"can":[45,114],"create":[46],"contention.":[48],"Specifically,":[49],"is":[52,101],"memory":[53,67],"I/O":[55],"intensive,":[56],"placing":[57],"it":[58],"in":[59,119],"direct":[60],"conflict":[61],"inference,":[64,124],"which":[65],"demands":[66],"for":[68,73],"KV":[69],"cache":[70],"compute":[72],"higher":[74],"throughput.":[75],"We":[76],"present":[77],"VectorLiterAG,":[78],"a":[79],"latency-aware":[80],"RAG":[81],"serving":[82],"system":[83,121],"explicitly":[85],"orchestrates":[86],"data":[87],"placement":[88],"execution":[90,127],"across":[91,149,205],"retrieval":[92,112],"inference":[94],"meet":[96],"strict":[97],"end-to-end":[98],"SLOs.":[99],"VectorLiterAG":[100,142],"driven":[102],"by":[103,230],"access-pattern":[104],"analysis":[105],"performance":[107],"estimation":[108],"regulate":[110],"how":[111],"variability":[113],"be":[115],"mitigated":[116],"managed":[118],"SLO-compliant":[126,202],"under":[128,164],"skewed":[129,165],"dynamic":[131],"workloads.":[132],"By":[133],"jointly":[134],"modeling":[135],"query":[139],"hit-rate":[140],"distributions,":[141,182],"identifies":[143],"an":[144],"optimal":[145],"index":[146,171],"partitioning":[147],"point":[148],"CPU":[150],"minimizes":[154],"contention":[155],"stabilizes":[157],"batching":[158,184],"behavior,":[159],"thereby":[160],"maximizing":[161],"sustained":[162],"throughput":[163,187,229],"access":[166,189],"patterns.":[167],"A":[168],"low-overhead":[169],"online":[170],"update":[172],"mechanism":[173],"allows":[174],"VectorLiterag":[175],"continuously":[177],"adapt":[178],"evolving":[180],"request":[181,203],"preserving":[183],"efficiency":[185],"as":[188],"patterns":[190],"evolve.":[191],"Our":[192],"evaluations":[193],"demonstrate":[194],"VecTORLITERAG":[196,218],"consistently":[197],"expands":[198],"range":[200],"of":[201],"rate":[204],"all":[206],"tested":[207],"configurations.":[208],"Without":[209],"increasing":[210],"generation":[212],"or":[214],"requiring":[215],"additional":[216],"hardware,":[217],"outperforms":[219],"both":[220],"naive":[221],"existing":[223],"alternative":[224],"frameworks,":[225],"improving":[226],"attainable":[227],"SLO-bound":[228],"up":[231],"to$\\mathbf{1.":[232],"5}":[233],"\\times$.":[234]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-03-05T00:00:00"}