{"id":"https://openalex.org/W4415062175","doi":"https://doi.org/10.1145/3770854.3780192","title":"PilotANN: Memory-Bounded GPU Acceleration for Vector Search","display_name":"PilotANN: Memory-Bounded GPU Acceleration for Vector Search","publication_year":2026,"publication_date":"2026-04-20","ids":{"openalex":"https://openalex.org/W4415062175","doi":"https://doi.org/10.1145/3770854.3780192"},"language":"en","primary_location":{"id":"doi:10.1145/3770854.3780192","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3770854.3780192","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3770854.3780192","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5078185398","display_name":"Yuntao Gui","orcid":"https://orcid.org/0000-0002-3593-3447"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yuntao Gui","raw_affiliation_strings":["The Chinese University of Hong Kong, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0002-3593-3447","affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102854732","display_name":"Peiqi Yin","orcid":"https://orcid.org/0009-0008-4081-913X"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Peiqi Yin","raw_affiliation_strings":["The Chinese University of Hong Kong, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0009-0008-4081-913X","affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027255721","display_name":"Yan Xiao","orcid":"https://orcid.org/0000-0002-2563-083X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao Yan","raw_affiliation_strings":["Centre for Perceptual and Interactive Intelligence, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0002-2122-915X","affiliations":[{"raw_affiliation_string":"Centre for Perceptual and Interactive Intelligence, Hong Kong, Hong Kong","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007808758","display_name":"Chaorui Zhang","orcid":"https://orcid.org/0000-0002-1530-2584"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chaorui Zhang","raw_affiliation_strings":["Theory Lab, 2012 Labs of Huawei, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0002-1530-2584","affiliations":[{"raw_affiliation_string":"Theory Lab, 2012 Labs of Huawei, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101915841","display_name":"Weixi Zhang","orcid":"https://orcid.org/0009-0007-4769-2779"},"institutions":[{"id":"https://openalex.org/I2250955327","display_name":"Huawei Technologies (China)","ror":"https://ror.org/00cmhce21","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250955327"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weixi Zhang","raw_affiliation_strings":["Theory Lab, 2012 Labs of Huawei, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0009-0007-4769-2779","affiliations":[{"raw_affiliation_string":"Theory Lab, 2012 Labs of Huawei, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I2250955327"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016082884","display_name":"James Cheng","orcid":"https://orcid.org/0000-0001-6313-6288"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"James Cheng","raw_affiliation_strings":["The Chinese University of Hong Kong, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0001-6313-6288","affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.00926469,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"348","last_page":"358"},"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":0.9804999828338623,"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":0.9804999828338623,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.9627000093460083,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.9592000246047974,"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/tree-traversal","display_name":"Tree traversal","score":0.7817000150680542},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.7788000106811523},{"id":"https://openalex.org/keywords/graph-traversal","display_name":"Graph traversal","score":0.6711999773979187},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5615000128746033},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.46869999170303345},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.44850000739097595},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4392000138759613},{"id":"https://openalex.org/keywords/search-algorithm","display_name":"Search algorithm","score":0.4251999855041504},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.39570000767707825}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8104000091552734},{"id":"https://openalex.org/C140745168","wikidata":"https://www.wikidata.org/wiki/Q1210082","display_name":"Tree traversal","level":2,"score":0.7817000150680542},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.7788000106811523},{"id":"https://openalex.org/C96333769","wikidata":"https://www.wikidata.org/wiki/Q907955","display_name":"Graph traversal","level":3,"score":0.6711999773979187},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5615000128746033},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.48080000281333923},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.46869999170303345},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.44850000739097595},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4392000138759613},{"id":"https://openalex.org/C125583679","wikidata":"https://www.wikidata.org/wiki/Q755673","display_name":"Search algorithm","level":2,"score":0.4251999855041504},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.39570000767707825},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.3952000141143799},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.37790000438690186},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3702999949455261},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.3447999954223633},{"id":"https://openalex.org/C116738811","wikidata":"https://www.wikidata.org/wiki/Q608751","display_name":"Nearest neighbor search","level":2,"score":0.34119999408721924},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3310000002384186},{"id":"https://openalex.org/C106515295","wikidata":"https://www.wikidata.org/wiki/Q26806595","display_name":"Parallel processing","level":2,"score":0.31470000743865967},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.31189998984336853},{"id":"https://openalex.org/C113238511","wikidata":"https://www.wikidata.org/wiki/Q1071612","display_name":"k-nearest neighbors algorithm","level":2,"score":0.296999990940094},{"id":"https://openalex.org/C19889080","wikidata":"https://www.wikidata.org/wiki/Q2835852","display_name":"Beam search","level":3,"score":0.28870001435279846},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.2800999879837036},{"id":"https://openalex.org/C12267149","wikidata":"https://www.wikidata.org/wiki/Q282453","display_name":"Support vector machine","level":2,"score":0.26330000162124634},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.2581999897956848},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.25679999589920044},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.2556000053882599},{"id":"https://openalex.org/C2989134064","wikidata":"https://www.wikidata.org/wiki/Q288510","display_name":"Execution time","level":2,"score":0.25450000166893005},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2513999938964844},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3770854.3780192","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3770854.3780192","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2503.21206","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.21206","pdf_url":"https://arxiv.org/pdf/2503.21206","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2503.21206","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2503.21206","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.1145/3770854.3780192","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3770854.3780192","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Approximate":[0],"Nearest":[1],"Neighbor":[2],"Search":[3],"(ANNS)":[4],"has":[5],"become":[6],"fundamental":[7],"to":[8,44,124,152,156],"modern":[9],"deep":[10],"learning":[11],"applications,":[12],"having":[13],"gained":[14],"particular":[15],"prominence":[16],"through":[17],"its":[18],"integration":[19],"into":[20,90],"recent":[21],"generative":[22],"models":[23],"that":[24,69,136],"work":[25],"with":[26],"increasingly":[27],"complex":[28],"datasets":[29,154],"and":[30,75,97,111,149],"higher":[31],"vector":[32],"dimensions.":[33],"Existing":[34],"CPU-only":[35],"solutions,":[36],"even":[37],"the":[38,86,161],"most":[39],"efficient":[40],"graph-based":[41,67],"ones,":[42],"struggle":[43],"meet":[45],"these":[46],"growing":[47],"computational":[48,99],"demands,":[49],"while":[50,129],"GPU-only":[51],"solutions":[52],"face":[53],"memory":[54],"constraints.":[55],"As":[56],"a":[57,62],"solution,":[58],"we":[59,119],"propose":[60],"PilotANN,":[61],"hybrid":[63],"CPU-GPU":[64],"system":[65],"for":[66],"ANNS":[68],"utilizes":[70],"both":[71],"CPU's":[72],"abundant":[73],"RAM":[74],"GPU's":[76],"parallel":[77],"processing":[78],"capabilities.":[79],"Our":[80],"key":[81],"innovation":[82],"lies":[83],"in":[84,143],"decomposing":[85],"top-k":[87],"search":[88,114,126],"process":[89],"three":[91],"complementary":[92],"stages":[93],"of":[94],"increasing":[95],"precision":[96],"decreasing":[98],"cost:":[100],"(i)":[101],"GPU-accelerated":[102],"subgraph":[103],"traversal":[104],"using":[105,115],"SVD-reduced":[106],"vectors;":[107],"(ii)":[108],"CPU":[109],"refinement;":[110],"(iii)":[112],"precise":[113],"complete":[116],"vectors.":[117],"Furthermore,":[118],"introduce":[120],"fast":[121],"entry":[122],"selection":[123],"improve":[125],"starting":[127],"points":[128],"maximizing":[130],"GPU":[131,162],"utilization.":[132],"Experimental":[133],"results":[134],"demonstrate":[135],"PilotANN":[137],"achieves":[138],"3.9":[139],"--":[140],"5.4\u00d7":[141],"speedup":[142],"throughput":[144],"on":[145],"100-million":[146],"scale":[147],"datasets,":[148],"is":[150],"able":[151],"handle":[153],"up":[155],"12":[157],"\u00d7":[158],"larger":[159],"than":[160],"memory.":[163]},"counts_by_year":[],"updated_date":"2026-08-05T07:39:15.569665","created_date":"2025-10-11T00:00:00"}