{"id":"https://openalex.org/W7109225768","doi":"https://doi.org/10.1145/3769799","title":"High-Throughput, Cost-Effective Billion-Scale Vector Search with a Single GPU","display_name":"High-Throughput, Cost-Effective Billion-Scale Vector Search with a Single GPU","publication_year":2025,"publication_date":"2025-12-04","ids":{"openalex":"https://openalex.org/W7109225768","doi":"https://doi.org/10.1145/3769799"},"language":"en","primary_location":{"id":"doi:10.1145/3769799","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3769799","pdf_url":null,"source":{"id":"https://openalex.org/S4387289859","display_name":"Proceedings of the ACM on Management of Data","issn_l":"2836-6573","issn":["2836-6573"],"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":"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 ACM on Management of Data","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1145/3769799","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Haodi Jiang","orcid":"https://orcid.org/0009-0007-2347-1680"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Haodi Jiang","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Hao Guo","orcid":"https://orcid.org/0009-0002-4819-5786"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Guo","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Minhui Xie","orcid":"https://orcid.org/0000-0001-6684-8336"},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Minhui Xie","raw_affiliation_strings":["Renmin University of China, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Renmin University of China, Beijing, China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jiwu Shu","orcid":"https://orcid.org/0000-0002-7362-2789"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiwu Shu","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":null,"display_name":"Youyou Lu","orcid":"https://orcid.org/0000-0002-6214-5390"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Youyou Lu","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.61161195,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"3","issue":"6","first_page":"1","last_page":"27"},"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":0.3734999895095825,"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":0.3734999895095825,"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":0.2660999894142151,"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/T12326","display_name":"Network Packet Processing and Optimization","score":0.11900000274181366,"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/bottleneck","display_name":"Bottleneck","score":0.7121999859809875},{"id":"https://openalex.org/keywords/pci-express","display_name":"PCI Express","score":0.6579999923706055},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.45329999923706055},{"id":"https://openalex.org/keywords/concurrency","display_name":"Concurrency","score":0.4129999876022339},{"id":"https://openalex.org/keywords/search-algorithm","display_name":"Search algorithm","score":0.3901999890804291},{"id":"https://openalex.org/keywords/search-engine","display_name":"Search engine","score":0.373199999332428},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.36320000886917114},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.3452000021934509}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8348000049591064},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.7121999859809875},{"id":"https://openalex.org/C64270927","wikidata":"https://www.wikidata.org/wiki/Q206924","display_name":"PCI Express","level":3,"score":0.6579999923706055},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.625},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.45329999923706055},{"id":"https://openalex.org/C193702766","wikidata":"https://www.wikidata.org/wiki/Q1414548","display_name":"Concurrency","level":2,"score":0.4129999876022339},{"id":"https://openalex.org/C125583679","wikidata":"https://www.wikidata.org/wiki/Q755673","display_name":"Search algorithm","level":2,"score":0.3901999890804291},{"id":"https://openalex.org/C97854310","wikidata":"https://www.wikidata.org/wiki/Q19541","display_name":"Search engine","level":2,"score":0.373199999332428},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.36320000886917114},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.3452000021934509},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.3434000015258789},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.32170000672340393},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.31299999356269836},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.30730000138282776},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3005000054836273},{"id":"https://openalex.org/C75165309","wikidata":"https://www.wikidata.org/wiki/Q2258979","display_name":"Search engine indexing","level":2,"score":0.28220000863075256},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.2757999897003174},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2750000059604645},{"id":"https://openalex.org/C12267149","wikidata":"https://www.wikidata.org/wiki/Q282453","display_name":"Support vector machine","level":2,"score":0.27459999918937683},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C150495011","wikidata":"https://www.wikidata.org/wiki/Q128392","display_name":"Concurrent computing","level":2,"score":0.25619998574256897},{"id":"https://openalex.org/C2777187653","wikidata":"https://www.wikidata.org/wiki/Q975106","display_name":"MPSoC","level":3,"score":0.2540999948978424},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.2508000135421753},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3769799","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3769799","pdf_url":null,"source":{"id":"https://openalex.org/S4387289859","display_name":"Proceedings of the ACM on Management of Data","issn_l":"2836-6573","issn":["2836-6573"],"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":"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 ACM on Management of Data","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1145/3769799","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3769799","pdf_url":null,"source":{"id":"https://openalex.org/S4387289859","display_name":"Proceedings of the ACM on Management of Data","issn_l":"2836-6573","issn":["2836-6573"],"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":"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 ACM on Management of Data","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W1970319631","https://openalex.org/W2059966434","https://openalex.org/W2086179657","https://openalex.org/W2124509324","https://openalex.org/W2132234208","https://openalex.org/W2427312773","https://openalex.org/W2468923260","https://openalex.org/W2797054769","https://openalex.org/W2963213349","https://openalex.org/W2963284996","https://openalex.org/W2984806015","https://openalex.org/W2992678377","https://openalex.org/W2998702515","https://openalex.org/W3028864969","https://openalex.org/W3036320503","https://openalex.org/W3093834826","https://openalex.org/W3166125679","https://openalex.org/W4318541614","https://openalex.org/W4387321071","https://openalex.org/W4390647563","https://openalex.org/W4396813431","https://openalex.org/W4399175194","https://openalex.org/W4400909749","https://openalex.org/W4401176373","https://openalex.org/W4407357331","https://openalex.org/W4407366079","https://openalex.org/W6922057760"],"related_works":[],"abstract_inverted_index":{"Approximate":[0],"nearest":[1],"neighbor":[2],"search":[3,17,60,111],"(ANNS)":[4],"is":[5,129],"broadly":[6],"adopted":[7],"in":[8,20,85,134],"numerous":[9],"scenarios.":[10],"Real-world":[11],"applications":[12],"seek":[13],"efficient":[14],"ways":[15],"to":[16,30,79,100,117],"billion-scale":[18,57],"vectors":[19],"high":[21,63],"throughput.":[22],"On-SSD":[23],"graph-based":[24,58],"ANNS":[25,51,119],"systems":[26],"have":[27],"the":[28,35,81,86,102,107],"opportunity":[29],"achieve":[31,68],"this":[32,44],"goal,":[33],"but":[34],"limited":[36],"CPU":[37],"computing":[38],"power":[39],"becomes":[40],"a":[41,48,56],"bottleneck.":[42],"In":[43],"paper,":[45],"we":[46],"propose":[47],"GPU-centric,":[49],"CPU-assisted":[50,98],"architecture":[52],"and":[53,65,95,128],"design":[54],"GustANN,":[55],"vector":[59],"system":[61],"for":[62,93,112],"throughput":[64],"cost-effectiveness.":[66],"We":[67],"these":[69],"goals":[70],"with":[71],"three":[72],"techniques:":[73],"(1)":[74],"memory-efficient":[75],"GPU":[76,82,94],"kernels":[77],"optimized":[78],"minimize":[80],"memory":[83],"usage":[84],"graph":[87],"search,":[88],"which":[89],"allows":[90],"higher":[91,126],"concurrency":[92],"SSD;":[96],"(2)":[97],"transfer":[99],"address":[101],"PCIe":[103],"bandwidth":[104],"bottleneck":[105],"on":[106],"GPU-side;":[108],"(3)":[109],"pivot":[110],"inter-SSD":[113],"load":[114],"balancing.":[115],"Compared":[116],"existing":[118],"systems,":[120],"GustANN":[121],"achieves":[122],"at":[123],"least":[124],"2.50\u00d7":[125],"throughput,":[127],"2.62\u00d7":[130],"more":[131],"cost-effective":[132],"(measured":[133],"/QPS).":[135]},"counts_by_year":[],"updated_date":"2025-12-06T23:14:57.273132","created_date":"2025-12-06T00:00:00"}