{"id":"https://openalex.org/W4415003901","doi":"https://doi.org/10.1145/3725783.3764393","title":"TARDIS: A GPU-Centric KV Cache Service for Efficient LLM Inference","display_name":"TARDIS: A GPU-Centric KV Cache Service for Efficient LLM Inference","publication_year":2025,"publication_date":"2025-10-09","ids":{"openalex":"https://openalex.org/W4415003901","doi":"https://doi.org/10.1145/3725783.3764393"},"language":"en","primary_location":{"id":"doi:10.1145/3725783.3764393","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3725783.3764393","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 16th ACM SIGOPS Asia-Pacific Workshop on Systems","raw_type":"proceedings-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/A5100708728","display_name":"Yifan Hu","orcid":"https://orcid.org/0009-0004-9393-5004"},"institutions":[{"id":"https://openalex.org/I191208505","display_name":"Xiamen University","ror":"https://ror.org/00mcjh785","country_code":"CN","type":"education","lineage":["https://openalex.org/I191208505"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yifan Hu","raw_affiliation_strings":["Xiamen University, Xiamen, Fujian, China"],"affiliations":[{"raw_affiliation_string":"Xiamen University, Xiamen, Fujian, China","institution_ids":["https://openalex.org/I191208505"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113911724","display_name":"Shi Qiu","orcid":"https://orcid.org/0000-0003-0821-0646"},"institutions":[{"id":"https://openalex.org/I191208505","display_name":"Xiamen University","ror":"https://ror.org/00mcjh785","country_code":"CN","type":"education","lineage":["https://openalex.org/I191208505"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shi Qiu","raw_affiliation_strings":["Xiamen University, Xiamen, Fujian, China"],"affiliations":[{"raw_affiliation_string":"Xiamen University, Xiamen, Fujian, China","institution_ids":["https://openalex.org/I191208505"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109440354","display_name":"Jiuming Yan","orcid":"https://orcid.org/0009-0009-4279-9395"},"institutions":[{"id":"https://openalex.org/I191208505","display_name":"Xiamen University","ror":"https://ror.org/00mcjh785","country_code":"CN","type":"education","lineage":["https://openalex.org/I191208505"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianqin Yan","raw_affiliation_strings":["Xiamen University, Xiamen, Fujian, China"],"affiliations":[{"raw_affiliation_string":"Xiamen University, Xiamen, Fujian, China","institution_ids":["https://openalex.org/I191208505"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112413246","display_name":"Hao Chen","orcid":"https://orcid.org/0009-0001-7141-5903"},"institutions":[{"id":"https://openalex.org/I191208505","display_name":"Xiamen University","ror":"https://ror.org/00mcjh785","country_code":"CN","type":"education","lineage":["https://openalex.org/I191208505"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Chen","raw_affiliation_strings":["Xiamen University, Xiamen, Fujian, China"],"affiliations":[{"raw_affiliation_string":"Xiamen University, Xiamen, Fujian, China","institution_ids":["https://openalex.org/I191208505"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072725991","display_name":"X. Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xintao Wang","raw_affiliation_strings":["Shanghai Jiao Tong University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101583971","display_name":"Lu Tang","orcid":"https://orcid.org/0000-0003-2923-6247"},"institutions":[{"id":"https://openalex.org/I191208505","display_name":"Xiamen University","ror":"https://ror.org/00mcjh785","country_code":"CN","type":"education","lineage":["https://openalex.org/I191208505"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tang Lu","raw_affiliation_strings":["Xiamen University, Xiamen, Fujian, China"],"affiliations":[{"raw_affiliation_string":"Xiamen University, Xiamen, Fujian, China","institution_ids":["https://openalex.org/I191208505"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101490654","display_name":"Guangtao Xue","orcid":"https://orcid.org/0000-0002-1617-3593"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guangtao Xue","raw_affiliation_strings":["SJTU, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"SJTU, Shanghai, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100395351","display_name":"Yiming Zhang","orcid":"https://orcid.org/0000-0001-6450-8485"},"institutions":[{"id":"https://openalex.org/I4210160814","display_name":"Risun (China)","ror":"https://ror.org/055vcsm02","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210160814"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yiming Zhang","raw_affiliation_strings":["XMU &amp; SJTU, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"XMU &amp; SJTU, Shanghai, China","institution_ids":["https://openalex.org/I4210160814"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5100708728"],"corresponding_institution_ids":["https://openalex.org/I191208505"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.33239611,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"46","last_page":"53"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9991000294685364,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9991000294685364,"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/T11269","display_name":"Algorithms and Data Compression","score":0.9977999925613403,"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"}},{"id":"https://openalex.org/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.9800999760627747,"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/cache","display_name":"Cache","score":0.7968999743461609},{"id":"https://openalex.org/keywords/asynchronous-communication","display_name":"Asynchronous communication","score":0.5752000212669373},{"id":"https://openalex.org/keywords/cache-algorithms","display_name":"Cache algorithms","score":0.507099986076355},{"id":"https://openalex.org/keywords/cache-invalidation","display_name":"Cache invalidation","score":0.5019999742507935},{"id":"https://openalex.org/keywords/synchronization","display_name":"Synchronization (alternating current)","score":0.4629000127315521},{"id":"https://openalex.org/keywords/smart-cache","display_name":"Smart Cache","score":0.45260000228881836},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.4496000111103058},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4165000021457672}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8366000056266785},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.7968999743461609},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6486999988555908},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.5752000212669373},{"id":"https://openalex.org/C38556500","wikidata":"https://www.wikidata.org/wiki/Q13404475","display_name":"Cache algorithms","level":4,"score":0.507099986076355},{"id":"https://openalex.org/C25536678","wikidata":"https://www.wikidata.org/wiki/Q5015977","display_name":"Cache invalidation","level":5,"score":0.5019999742507935},{"id":"https://openalex.org/C2778562939","wikidata":"https://www.wikidata.org/wiki/Q1298791","display_name":"Synchronization (alternating current)","level":3,"score":0.4629000127315521},{"id":"https://openalex.org/C167713795","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"Smart Cache","level":5,"score":0.45260000228881836},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.4496000111103058},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4165000021457672},{"id":"https://openalex.org/C189783530","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"CPU cache","level":3,"score":0.3718000054359436},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.3393999934196472},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.334199994802475},{"id":"https://openalex.org/C2780378061","wikidata":"https://www.wikidata.org/wiki/Q25351891","display_name":"Service (business)","level":2,"score":0.32659998536109924},{"id":"https://openalex.org/C201148951","wikidata":"https://www.wikidata.org/wiki/Q5015976","display_name":"Cache coloring","level":4,"score":0.31929999589920044},{"id":"https://openalex.org/C59687516","wikidata":"https://www.wikidata.org/wiki/Q5015938","display_name":"Cache-oblivious algorithm","level":5,"score":0.30399999022483826},{"id":"https://openalex.org/C113166858","wikidata":"https://www.wikidata.org/wiki/Q5015981","display_name":"Cache pollution","level":5,"score":0.3000999987125397},{"id":"https://openalex.org/C1793878","wikidata":"https://www.wikidata.org/wiki/Q1153762","display_name":"Out-of-order execution","level":2,"score":0.28949999809265137},{"id":"https://openalex.org/C120936851","wikidata":"https://www.wikidata.org/wiki/Q1408065","display_name":"MESI protocol","level":5,"score":0.27390000224113464},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.251800000667572}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3725783.3764393","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3725783.3764393","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 16th ACM SIGOPS Asia-Pacific Workshop on Systems","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":8,"referenced_works":["https://openalex.org/W2734941459","https://openalex.org/W4205983429","https://openalex.org/W4318541614","https://openalex.org/W4387321091","https://openalex.org/W4391463154","https://openalex.org/W4401176373","https://openalex.org/W4408844835","https://openalex.org/W4408848702"],"related_works":[],"abstract_inverted_index":{"Key-value":[0],"(KV)":[1],"cache":[2,31],"is":[3],"a":[4,24],"crucial":[5],"optimization":[6],"for":[7],"large":[8],"language":[9],"model":[10],"(LLM)":[11],"serving,":[12],"particularly":[13],"in":[14],"long-context":[15],"inference":[16],"scenarios.":[17],"While":[18],"existing":[19],"KV":[20,30,61],"stores":[21],"suffer":[22],"from":[23],"fundamental":[25],"mismatch":[26],"between":[27,53],"the":[28,34,38,43,51,64],"CPU-centric":[29],"storage":[32],"and":[33,49,55],"GPU-accelerated":[35],"computation:":[36],"(1)":[37],"low-parallelism":[39],"CPUs":[40,54],"cannot":[41],"satisfy":[42],"fine-grained,":[44],"highly":[45],"parallel":[46],"I/O":[47],"demand,":[48],"(2)":[50],"synchronization":[52],"GPUs":[56],"introduced":[57],"by":[58],"layerwise":[59],"asynchronous":[60],"operations":[62],"prevents":[63],"adoption":[65],"of":[66],"low-level":[67],"optimizations":[68],"(such":[69],"as":[70],"CUDA":[71],"Graph).":[72]},"counts_by_year":[],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}
