{"id":"https://openalex.org/W7151489934","doi":"https://doi.org/10.1145/3786706","title":"VecFlow-Chamfer: A GPU-based Data Management System for High-Performance Multi-Vector Search on Superchips","display_name":"VecFlow-Chamfer: A GPU-based Data Management System for High-Performance Multi-Vector Search on Superchips","publication_year":2026,"publication_date":"2026-04-02","ids":{"openalex":"https://openalex.org/W7151489934","doi":"https://doi.org/10.1145/3786706"},"language":"en","primary_location":{"id":"doi:10.1145/3786706","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3786706","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","license_id":"https://openalex.org/licenses/cc-by","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/10.1145/3786706","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5065967981","display_name":"Chenghao Mo","orcid":"https://orcid.org/0009-0003-9860-6325"},"institutions":[{"id":"https://openalex.org/I183874917","display_name":"Urbana University","ror":"https://ror.org/04kp3hw27","country_code":"US","type":"education","lineage":["https://openalex.org/I183874917"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chenghao Mo","raw_affiliation_strings":["SSAIL Lab, UIUC, Urbana, IL, USA"],"raw_orcid":"https://orcid.org/0009-0003-9860-6325","affiliations":[{"raw_affiliation_string":"SSAIL Lab, UIUC, Urbana, IL, USA","institution_ids":["https://openalex.org/I183874917"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057388897","display_name":"Ben Karsin","orcid":"https://orcid.org/0000-0003-4011-8903"},"institutions":[{"id":"https://openalex.org/I1304085615","display_name":"Nvidia (United Kingdom)","ror":"https://ror.org/02kr42612","country_code":"GB","type":"company","lineage":["https://openalex.org/I1304085615","https://openalex.org/I4210127875"]},{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Ben Karsin","raw_affiliation_strings":["Nvidia, Honolulu, USA"],"raw_orcid":"https://orcid.org/0000-0003-4011-8903","affiliations":[{"raw_affiliation_string":"Nvidia, Honolulu, USA","institution_ids":["https://openalex.org/I1304085615","https://openalex.org/I4210127875"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133121399","display_name":"Philip Adams","orcid":null},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Philip Adams","raw_affiliation_strings":["Microsoft, Redmond, USA"],"raw_orcid":"https://orcid.org/0009-0000-6365-1853","affiliations":[{"raw_affiliation_string":"Microsoft, Redmond, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077768924","display_name":"Minjia Zhang","orcid":"https://orcid.org/0000-0002-8165-166X"},"institutions":[{"id":"https://openalex.org/I183874917","display_name":"Urbana University","ror":"https://ror.org/04kp3hw27","country_code":"US","type":"education","lineage":["https://openalex.org/I183874917"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Minjia Zhang","raw_affiliation_strings":["SSAIL Lab, UIUC, Urbana, IL, USA"],"raw_orcid":"https://orcid.org/0000-0002-8165-166X","affiliations":[{"raw_affiliation_string":"SSAIL Lab, UIUC, Urbana, IL, USA","institution_ids":["https://openalex.org/I183874917"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"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":0.44850445,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"4","issue":"1","first_page":"1","last_page":"26"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.20090000331401825,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.20090000331401825,"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"}},{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.16949999332427979,"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/T12292","display_name":"Graph Theory and Algorithms","score":0.10109999775886536,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6758999824523926},{"id":"https://openalex.org/keywords/search-engine-indexing","display_name":"Search engine indexing","score":0.6467999815940857},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.4129999876022339},{"id":"https://openalex.org/keywords/nearest-neighbor-search","display_name":"Nearest neighbor search","score":0.4016999900341034},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.37049999833106995},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.3677000105381012},{"id":"https://openalex.org/keywords/data-management","display_name":"Data management","score":0.3425999879837036},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.3407000005245209},{"id":"https://openalex.org/keywords/relevance","display_name":"Relevance (law)","score":0.33309999108314514}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.835099995136261},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6758999824523926},{"id":"https://openalex.org/C75165309","wikidata":"https://www.wikidata.org/wiki/Q2258979","display_name":"Search engine indexing","level":2,"score":0.6467999815940857},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.5040000081062317},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.4478999972343445},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4129999876022339},{"id":"https://openalex.org/C116738811","wikidata":"https://www.wikidata.org/wiki/Q608751","display_name":"Nearest neighbor search","level":2,"score":0.4016999900341034},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.37049999833106995},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.3677000105381012},{"id":"https://openalex.org/C1668388","wikidata":"https://www.wikidata.org/wiki/Q1149776","display_name":"Data management","level":2,"score":0.3425999879837036},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.3407000005245209},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.33309999108314514},{"id":"https://openalex.org/C2780762811","wikidata":"https://www.wikidata.org/wiki/Q1784941","display_name":"Cosine similarity","level":3,"score":0.33219999074935913},{"id":"https://openalex.org/C551230270","wikidata":"https://www.wikidata.org/wiki/Q4368942","display_name":"Data retrieval","level":2,"score":0.3278000056743622},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.31630000472068787},{"id":"https://openalex.org/C47487241","wikidata":"https://www.wikidata.org/wiki/Q5227230","display_name":"Data access","level":2,"score":0.28940001130104065},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C194739806","wikidata":"https://www.wikidata.org/wiki/Q66221","display_name":"Computer data storage","level":2,"score":0.28380000591278076},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.28279998898506165},{"id":"https://openalex.org/C82687282","wikidata":"https://www.wikidata.org/wiki/Q66221","display_name":"Auxiliary memory","level":2,"score":0.274399995803833},{"id":"https://openalex.org/C68859911","wikidata":"https://www.wikidata.org/wiki/Q1503724","display_name":"Pattern matching","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C97854310","wikidata":"https://www.wikidata.org/wiki/Q19541","display_name":"Search engine","level":2,"score":0.26350000500679016},{"id":"https://openalex.org/C166423231","wikidata":"https://www.wikidata.org/wiki/Q1891170","display_name":"Semantic search","level":3,"score":0.26080000400543213},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2603999972343445},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.2567000091075897},{"id":"https://openalex.org/C81669768","wikidata":"https://www.wikidata.org/wiki/Q2359161","display_name":"Precision and recall","level":2,"score":0.25529998540878296},{"id":"https://openalex.org/C198783460","wikidata":"https://www.wikidata.org/wiki/Q629173","display_name":"Management system","level":2,"score":0.2540000081062317},{"id":"https://openalex.org/C99138194","wikidata":"https://www.wikidata.org/wiki/Q183427","display_name":"Hash function","level":2,"score":0.2508000135421753}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3786706","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3786706","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","license_id":"https://openalex.org/licenses/cc-by","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:10.1145/3786706","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3786706","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","license_id":"https://openalex.org/licenses/cc-by","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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W145388748","https://openalex.org/W2119913432","https://openalex.org/W2512971201","https://openalex.org/W2905266130","https://openalex.org/W2912924812","https://openalex.org/W2984806015","https://openalex.org/W3021397474","https://openalex.org/W3136183693","https://openalex.org/W3199493219","https://openalex.org/W3217305727","https://openalex.org/W4306317212","https://openalex.org/W4390962868","https://openalex.org/W4412377187","https://openalex.org/W7133211346"],"related_works":[],"abstract_inverted_index":{"Many":[0],"emerging":[1],"AI":[2],"applications":[3],"demand":[4],"retrieval":[5,75],"systems":[6,55,206],"that":[7,92,146,167],"go":[8],"beyond":[9],"document-level":[10],"relevance":[11],"and":[12,30,45,63,77,95,131,159,200,209],"capture":[13],"token-level":[14,26,52],"semantics.":[15],"Multi-vector":[16],"search":[17,98,122,187],"addresses":[18],"this":[19,105],"need":[20],"through":[21,59,106],"fine-grained":[22,124],"semantic":[23],"matching":[24],"between":[25],"embeddings":[27,174,195],"of":[28,73,109,154,156,190,192],"queries":[29],"documents.":[31],"However,":[32],"it":[33],"introduces":[34],"significant":[35],"system":[36,91],"challenges,":[37],"including":[38],"compute-intensive":[39],"set-to-set":[40],"scoring,":[41],"complex":[42],"candidate":[43,134,157],"filtering,":[44],"high":[46,198],"memory":[47],"overhead":[48],"from":[49,79],"storing":[50],"dense":[51],"embeddings.":[53],"Prior":[54],"mitigate":[56],"these":[57,180],"challenges":[58],"GPU-based":[60,87],"similarity":[61],"calculation":[62],"indexing":[64],"structures":[65],"(e.g.,":[66],"IVFPQ-GPU)":[67],"but":[68],"often":[69],"at":[70],"the":[71],"cost":[72],"reduced":[74],"accuracy":[76],"suffer":[78],"low":[80,201],"GPU":[81,138,144],"utilization.":[82],"We":[83],"present":[84],"\\name,":[85],"a":[86,107,115,142,162],"vector":[88,164],"data":[89],"management":[90],"enables":[93,147],"low-latency":[94,170],"high-recall":[96],"multi-vector":[97,121,186],"on":[99],"modern":[100],"Superchip":[101],"architectures.":[102],"\\name":[103,183,211],"achieves":[104,212],"combination":[108],"three":[110],"novel":[111],"optimizations:":[112],"(1)":[113],"\\maxivf,":[114],"GPU-native,":[116],"compression-free":[117],"index":[118,129],"tailored":[119],"for":[120],"with":[123,196],"anchor":[125],"vectors":[126],"enabling":[127],"scalable":[128],"construction":[130],"low-latency,":[132],"high-accuracy":[133],"generation":[135],"via":[136],"CAGRA-based":[137],"routing;":[139],"(2)":[140],"\\chamferkernel,":[141],"highly-optimized":[143],"kernel":[145],"single-digit":[148],"millisecond":[149],"Chamfer":[150],"scoring":[151],"over":[152,188],"tens":[153],"thousands":[155],"documents;":[158],"(3)":[160],"\\zerocomp,":[161],"tiered":[163],"storage":[165],"layer":[166],"supports":[168],"on-demand,":[169],"access":[171],"to":[172,184,204],"full-precision":[173],"across":[175],"Grace-Hopper":[176],"NVLink-C2C":[177],"interconnects.":[178],"Together,":[179],"techniques":[181],"enable":[182],"perform":[185],"hundreds":[189],"millions":[191],"document":[193],"token":[194],"unprecedented":[197],"recall":[199],"latency.":[202],"Compared":[203],"state-of-the-art":[205],"like":[207],"PLAID":[208],"MUVERA,":[210],"an":[213],"order-of-magnitude":[214],"lower":[215],"latency":[216],"while":[217],"significantly":[218],"improving":[219],"recall.":[220]},"counts_by_year":[],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2026-04-08T00:00:00"}