{"id":"https://openalex.org/W4411486516","doi":"https://doi.org/10.1145/3695053.3731032","title":"In-Storage Acceleration of Retrieval Augmented Generation as a Service","display_name":"In-Storage Acceleration of Retrieval Augmented Generation as a Service","publication_year":2025,"publication_date":"2025-06-20","ids":{"openalex":"https://openalex.org/W4411486516","doi":"https://doi.org/10.1145/3695053.3731032"},"language":"en","primary_location":{"id":"doi:10.1145/3695053.3731032","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3695053.3731032","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3695053.3731032","source":null,"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd Annual International Symposium on Computer Architecture","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3695053.3731032","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5047989184","display_name":"Rohan Mahapatra","orcid":"https://orcid.org/0000-0002-2887-9761"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rohan Mahapatra","raw_affiliation_strings":["University of California San Diego, La Jolla, California, USA"],"raw_orcid":"https://orcid.org/0000-0002-2887-9761","affiliations":[{"raw_affiliation_string":"University of California San Diego, La Jolla, California, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5118513708","display_name":"Harsha Santhanam","orcid":null},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Harsha Santhanam","raw_affiliation_strings":["University of California San Diego, La Jolla, California, USA"],"raw_orcid":"https://orcid.org/0009-0008-1636-5430","affiliations":[{"raw_affiliation_string":"University of California San Diego, La Jolla, California, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5118513709","display_name":"Christopher Priebe","orcid":"https://orcid.org/0009-0002-4028-6448"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Christopher Priebe","raw_affiliation_strings":["University of California San Diego, La Jolla, California, USA"],"raw_orcid":"https://orcid.org/0009-0002-4028-6448","affiliations":[{"raw_affiliation_string":"University of California San Diego, La Jolla, California, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078385434","display_name":"Hanyang Xu","orcid":"https://orcid.org/0009-0000-0328-9610"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hanyang Xu","raw_affiliation_strings":["University of California San Diego, La Jolla, California, USA"],"raw_orcid":"https://orcid.org/0009-0000-0328-9610","affiliations":[{"raw_affiliation_string":"University of California San Diego, La Jolla, California, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5084514143","display_name":"Hadi Esmaeilzadeh","orcid":"https://orcid.org/0000-0002-8548-1039"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hadi Esmaeilzadeh","raw_affiliation_strings":["University of California San Diego, La Jolla, California, USA"],"raw_orcid":"https://orcid.org/0000-0002-8548-1039","affiliations":[{"raw_affiliation_string":"University of California San Diego, La Jolla, California, USA","institution_ids":["https://openalex.org/I36258959"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I36258959"],"apc_list":null,"apc_paid":null,"fwci":13.5279,"has_fulltext":true,"cited_by_count":8,"citation_normalized_percentile":{"value":0.99028622,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"450","last_page":"466"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11478","display_name":"Caching and Content Delivery","score":0.9991999864578247,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T11478","display_name":"Caching and Content Delivery","score":0.9991999864578247,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9984999895095825,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9955000281333923,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/acceleration","display_name":"Acceleration","score":0.6961606740951538},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6955066919326782},{"id":"https://openalex.org/keywords/service","display_name":"Service (business)","score":0.5310922861099243},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.09108752012252808},{"id":"https://openalex.org/keywords/business","display_name":"Business","score":0.08067604899406433}],"concepts":[{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.6961606740951538},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6955066919326782},{"id":"https://openalex.org/C2780378061","wikidata":"https://www.wikidata.org/wiki/Q25351891","display_name":"Service (business)","level":2,"score":0.5310922861099243},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.09108752012252808},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.08067604899406433},{"id":"https://openalex.org/C162853370","wikidata":"https://www.wikidata.org/wiki/Q39809","display_name":"Marketing","level":1,"score":0.0},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3695053.3731032","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3695053.3731032","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3695053.3731032","source":null,"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd Annual International Symposium on Computer Architecture","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3695053.3731032","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3695053.3731032","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3695053.3731032","source":null,"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd Annual International Symposium on Computer Architecture","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.7599999904632568,"display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G137895390","display_name":"Collaborative Research: SHF: Medium: Spatial Multi-Tenant Neural Acceleration for Next Generation Datacenters","funder_award_id":"2107598","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320332195","display_name":"Samsung","ror":"https://ror.org/04w3jy968"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411486516.pdf","grobid_xml":"https://content.openalex.org/works/W4411486516.grobid-xml"},"referenced_works_count":50,"referenced_works":["https://openalex.org/W1965869859","https://openalex.org/W2004376136","https://openalex.org/W2070525241","https://openalex.org/W2105102111","https://openalex.org/W2111006384","https://openalex.org/W2883730939","https://openalex.org/W2963265099","https://openalex.org/W2963284996","https://openalex.org/W2963469388","https://openalex.org/W2963514026","https://openalex.org/W2979937837","https://openalex.org/W3021397474","https://openalex.org/W3043776758","https://openalex.org/W3047848469","https://openalex.org/W3130240120","https://openalex.org/W3130716829","https://openalex.org/W3154280800","https://openalex.org/W3155487259","https://openalex.org/W3155807546","https://openalex.org/W3180230246","https://openalex.org/W3194782062","https://openalex.org/W3196923642","https://openalex.org/W3206453033","https://openalex.org/W4214559301","https://openalex.org/W4214821564","https://openalex.org/W4226000737","https://openalex.org/W4252076394","https://openalex.org/W4283692994","https://openalex.org/W4300865759","https://openalex.org/W4312285477","https://openalex.org/W4361001766","https://openalex.org/W4385573970","https://openalex.org/W4387321071","https://openalex.org/W4387846582","https://openalex.org/W4388778348","https://openalex.org/W4388994251","https://openalex.org/W4390647563","https://openalex.org/W4391098193","https://openalex.org/W4392597393","https://openalex.org/W4393407266","https://openalex.org/W4394998968","https://openalex.org/W4394998969","https://openalex.org/W4395020674","https://openalex.org/W4395020696","https://openalex.org/W4402671835","https://openalex.org/W4403600390","https://openalex.org/W4404534210","https://openalex.org/W4404782930","https://openalex.org/W4407197060","https://openalex.org/W4407355496"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Retrieval-augmented":[0],"generation":[1,27,204,208],"(RAG)":[2],"services":[3,89],"are":[4,90],"rapidly":[5],"gaining":[6],"adoption":[7],"in":[8,86,223],"enterprise":[9],"settings":[10],"as":[11,124],"they":[12],"combine":[13],"information":[14,39],"retrieval":[15,229],"systems":[16],"(e.g.,":[17],"databases)":[18],"with":[19,37,234,240],"large":[20,50],"language":[21,138,199],"models":[22,43,200],"(LLMs)":[23],"to":[24,44,49,70,96,128,140,160,180,218,227],"enhance":[25,101],"response":[26],"and":[28,77,103,147,154,188,220,237],"reduce":[29],"hallucinations.By":[30],"augmenting":[31],"an":[32],"LLM's":[33,72],"fixed":[34],"pre-trained":[35],"knowledge":[36,51,73,99],"real-time":[38],"retrieval,":[40],"RAG":[41,63,88,183],"enables":[42],"effectively":[45],"extend":[46],"their":[47],"context":[48],"bases":[52],"by":[53],"selectively":[54],"retrieving":[55],"only":[56],"the":[57,65,71,83,110,118,125,177],"most":[58],"relevant":[59],"information.As":[60],"a":[61,136,170],"result,":[62],"provides":[64,176],"effect":[66],"of":[67,166,197],"dynamic":[68,185],"updates":[69],"without":[74],"requiring":[75],"expensive":[76],"time-consuming":[78],"retraining.While":[79],"some":[80],"deployments":[81],"keep":[82],"entire":[84],"database":[85],"memory,":[87],"increasingly":[91],"shifting":[92],"toward":[93],"persistent":[94,161],"storage":[95,236],"accommodate":[97],"ever-growing":[98],"bases,":[100],"utility,":[102],"improve":[104],"cost-efficiency.However,":[105],"this":[106,164],"transition":[107],"fundamentally":[108],"reshapes":[109],"system's":[111],"performance":[112],"profile:":[113],"empirical":[114],"analysis":[115],"reveals":[116],"that":[117,175],"Search":[119],"&":[120],"Retrieval":[121],"phase":[122,131],"emerges":[123],"dominant":[126],"contributor":[127],"end-to-end":[129,224],"latency.This":[130],"typically":[132],"involves":[133],"(1)":[134],"running":[135],"smaller":[137,198],"model":[139],"generate":[141],"query":[142,202],"embeddings,":[143],"(2)":[144],"executing":[145],"similarity":[146],"relevance":[148],"checks":[149],"over":[150],"varying":[151,189],"data":[152,186],"structures,":[153,187],"(3)":[155],"performing":[156],"frequent,":[157],"long-latency":[158],"accesses":[159],"storage.To":[162],"address":[163],"triad":[165],"challenges,":[167],"we":[168],"propose":[169],"metamorphic":[171],"in-storage":[172,195],"accelerator":[173],"architecture":[174,192],"necessary":[178],"programmability":[179],"support":[181],"diverse":[182],"algorithms,":[184],"computational":[190],"patterns.The":[191],"also":[193],"supports":[194],"execution":[196],"for":[201],"embedding":[203],"while":[205],"final":[206],"LLM":[207],"is":[209],"executed":[210],"on":[211],"DGX":[212],"A100":[213,238],"systems.Experimental":[214],"results":[215],"show":[216],"up":[217],"4.3":[219],"1.5":[221],"improvement":[222],"throughput":[225],"compared":[226],"conventional":[228],"pipelines":[230],"using":[231],"Xeon":[232],"CPUs":[233],"NVMe":[235],"GPUs":[239],"DRAM,":[241],"respectively.":[242]},"counts_by_year":[{"year":2026,"cited_by_count":7},{"year":2025,"cited_by_count":1}],"updated_date":"2026-08-11T07:18:39.950985","created_date":"2025-10-10T00:00:00"}
