{"id":"https://openalex.org/W4407355496","doi":"https://doi.org/10.14778/3696435.3696439","title":"Chameleon: A Heterogeneous and Disaggregated Accelerator System for Retrieval-Augmented Language Models","display_name":"Chameleon: A Heterogeneous and Disaggregated Accelerator System for Retrieval-Augmented Language Models","publication_year":2024,"publication_date":"2024-09-01","ids":{"openalex":"https://openalex.org/W4407355496","doi":"https://doi.org/10.14778/3696435.3696439"},"language":"en","primary_location":{"id":"doi:10.14778/3696435.3696439","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3696435.3696439","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the VLDB Endowment","raw_type":"journal-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/A5031344912","display_name":"Wenqi Jiang","orcid":"https://orcid.org/0000-0003-3895-7943"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Wenqi Jiang","raw_affiliation_strings":["Systems Group, ETH Zurich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Systems Group, ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100918465","display_name":"Marco Zeller","orcid":null},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Marco Zeller","raw_affiliation_strings":["Systems Group, ETH Zurich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Systems Group, ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039817377","display_name":"Roger Waleffe","orcid":"https://orcid.org/0000-0002-3795-4997"},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Roger Waleffe","raw_affiliation_strings":["University of Wisconsin Madison"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wisconsin Madison","institution_ids":["https://openalex.org/I135310074"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026990786","display_name":"Torsten Hoefler","orcid":"https://orcid.org/0000-0002-1333-9797"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Torsten Hoefler","raw_affiliation_strings":["SPCL, ETH Zurich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SPCL, ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5103144919","display_name":"Gustavo Alonso","orcid":"https://orcid.org/0000-0002-4396-6695"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Gustavo Alonso","raw_affiliation_strings":["Systems Group, ETH Zurich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Systems Group, ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.8972,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":{"value":0.92379571,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":"18","issue":"1","first_page":"42","last_page":"52"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9994000196456909,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9994000196456909,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9977999925613403,"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/T10028","display_name":"Topic Modeling","score":0.9976000189781189,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5433474779129028},{"id":"https://openalex.org/keywords/environmental-science","display_name":"Environmental science","score":0.3638821840286255}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5433474779129028},{"id":"https://openalex.org/C39432304","wikidata":"https://www.wikidata.org/wiki/Q188847","display_name":"Environmental science","level":0,"score":0.3638821840286255}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.14778/3696435.3696439","is_oa":false,"landing_page_url":"https://doi.org/10.14778/3696435.3696439","pdf_url":null,"source":{"id":"https://openalex.org/S4210226185","display_name":"Proceedings of the VLDB Endowment","issn_l":"2150-8097","issn":["2150-8097"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the VLDB Endowment","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":51,"referenced_works":["https://openalex.org/W642889137","https://openalex.org/W1502916507","https://openalex.org/W2011444829","https://openalex.org/W2021651899","https://openalex.org/W2042120213","https://openalex.org/W2062153469","https://openalex.org/W2077815765","https://openalex.org/W2086179657","https://openalex.org/W2096578865","https://openalex.org/W2102061396","https://openalex.org/W2124509324","https://openalex.org/W2161708191","https://openalex.org/W2162006472","https://openalex.org/W2248601845","https://openalex.org/W2292011317","https://openalex.org/W2441967103","https://openalex.org/W2460613287","https://openalex.org/W2734941459","https://openalex.org/W2799244653","https://openalex.org/W2906958661","https://openalex.org/W2952307259","https://openalex.org/W2963469388","https://openalex.org/W2969388332","https://openalex.org/W2998702515","https://openalex.org/W3027879771","https://openalex.org/W3029865833","https://openalex.org/W3032358677","https://openalex.org/W3037854022","https://openalex.org/W3085011441","https://openalex.org/W3103902563","https://openalex.org/W3129831491","https://openalex.org/W3157700644","https://openalex.org/W3174809957","https://openalex.org/W3175564083","https://openalex.org/W3197264354","https://openalex.org/W4200515834","https://openalex.org/W4285355789","https://openalex.org/W4301461621","https://openalex.org/W4306176157","https://openalex.org/W4380433185","https://openalex.org/W4381329135","https://openalex.org/W4381329139","https://openalex.org/W4381610063","https://openalex.org/W4385653226","https://openalex.org/W4387321071","https://openalex.org/W4387321091","https://openalex.org/W4389491850","https://openalex.org/W4398230026","https://openalex.org/W4399175194","https://openalex.org/W6629956336","https://openalex.org/W6779791889"],"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":{"A":[0],"Retrieval-Augmented":[1],"Language":[2],"Model":[3],"(RALM)":[4],"combines":[5],"a":[6,12,50,62],"large":[7],"language":[8],"model":[9],"(LLM)":[10],"with":[11,29,107],"vector":[13,58,84,96,152],"database":[14],"to":[15,87,105,119,130],"retrieve":[16],"context-specific":[17],"knowledge":[18],"during":[19],"text":[20],"generation.":[21],"This":[22],"strategy":[23],"facilitates":[24],"impressive":[25],"generation":[26],"quality":[27],"even":[28],"smaller":[30],"models,":[31],"thus":[32],"reducing":[33],"computational":[34],"demands":[35],"by":[36],"orders":[37],"of":[38,81],"magnitude.":[39],"To":[40],"serve":[41],"RALMs":[42],"efficiently":[43],"and":[44,57,73,83,101,124],"flexibly,":[45],"we":[46],"propose":[47],"Chameleon":[48,93,116],",":[49],"heterogeneous":[51,143],"accelerator":[52],"system":[53],"integrating":[54],"both":[55,71],"LLM":[56,82,103,148],"search":[59,85,97,153],"accelerators":[60,86,98,144],"in":[61,122,127,154],"disaggregated":[63],"architecture.":[64,134],"The":[65,135],"heterogeneity":[66],"ensures":[67],"efficient":[68],"serving":[69],"for":[70,141,145],"inference":[72,104,149],"retrieval,":[74],"while":[75],"the":[76,131,139],"disaggregation":[77],"allows":[78],"independent":[79],"scaling":[80],"fulfill":[88],"diverse":[89],"RALM":[90,156],"requirements.":[91],"Our":[92],"prototype":[94],"implements":[95],"on":[99,113],"FPGAs":[100],"assigns":[102],"GPUs,":[106],"CPUs":[108],"as":[109],"cluster":[110],"coordinators.":[111],"Evaluated":[112],"various":[114],"RALMs,":[115],"exhibits":[117],"up":[118],"2.16\u00d7":[120],"reduction":[121],"latency":[123],"3.18\u00d7":[125],"speedup":[126],"throughput":[128],"compared":[129],"hybrid":[132],"CPU-GPU":[133],"promising":[136],"results":[137],"pave":[138],"way":[140],"adopting":[142],"not":[146],"only":[147],"but":[150],"also":[151],"future":[155],"systems.":[157]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":9}],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-10T00:00:00"}