{"id":"https://openalex.org/W4416203787","doi":"https://doi.org/10.1145/3712285.3759777","title":"UpANNS: Enhancing Billion-Scale ANNS Efficiency with Real-World PIM Architecture","display_name":"UpANNS: Enhancing Billion-Scale ANNS Efficiency with Real-World PIM Architecture","publication_year":2025,"publication_date":"2025-11-12","ids":{"openalex":"https://openalex.org/W4416203787","doi":"https://doi.org/10.1145/3712285.3759777"},"language":null,"primary_location":{"id":"doi:10.1145/3712285.3759777","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3712285.3759777","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 International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3712285.3759777","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5113404328","display_name":"Sitian Chen","orcid":"https://orcid.org/0009-0008-7734-7754"},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Sitian Chen","raw_affiliation_strings":["Hong Kong Baptist University, Hong Kong, China"],"raw_orcid":"https://orcid.org/0009-0008-7734-7754","affiliations":[{"raw_affiliation_string":"Hong Kong Baptist University, Hong Kong, China","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102803813","display_name":"Amelie Chi Zhou","orcid":"https://orcid.org/0000-0002-9773-9332"},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Amelie Chi Zhou","raw_affiliation_strings":["Hong Kong Baptist University, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-9773-9332","affiliations":[{"raw_affiliation_string":"Hong Kong Baptist University, Hong Kong, China","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113227812","display_name":"Yinuo Shi","orcid":"https://orcid.org/0009-0005-8998-2411"},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yucheng Shi","raw_affiliation_strings":["Hong Kong Baptist University, Hong Kong, China"],"raw_orcid":"https://orcid.org/0009-0005-8998-2411","affiliations":[{"raw_affiliation_string":"Hong Kong Baptist University, Hong Kong, China","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026498734","display_name":"Yusen Li","orcid":"https://orcid.org/0000-0001-6623-350X"},"institutions":[{"id":"https://openalex.org/I205237279","display_name":"Nankai University","ror":"https://ror.org/01y1kjr75","country_code":"CN","type":"education","lineage":["https://openalex.org/I205237279"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yusen Li","raw_affiliation_strings":["Nankai University, Tianjin, China"],"raw_orcid":"https://orcid.org/0000-0001-6623-350X","affiliations":[{"raw_affiliation_string":"Nankai University, Tianjin, China","institution_ids":["https://openalex.org/I205237279"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071840581","display_name":"Xin Yao","orcid":"https://orcid.org/0000-0002-5780-7900"},"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":"Xin Yao","raw_affiliation_strings":["Huawei, Hong Kong, China"],"raw_orcid":"https://orcid.org/0000-0002-5780-7900","affiliations":[{"raw_affiliation_string":"Huawei, Hong Kong, China","institution_ids":["https://openalex.org/I2250955327"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"789","last_page":"804"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12292","display_name":"Graph Theory and Algorithms","score":0.26019999384880066,"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/T12292","display_name":"Graph Theory and Algorithms","score":0.26019999384880066,"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/T11106","display_name":"Data Management and Algorithms","score":0.1657000035047531,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.10689999908208847,"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":0.7533000111579895},{"id":"https://openalex.org/keywords/workload","display_name":"Workload","score":0.54339998960495},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5375999808311462},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.5293999910354614},{"id":"https://openalex.org/keywords/implementation","display_name":"Implementation","score":0.517799973487854},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.4410000145435333},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.4246000051498413},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.421999990940094},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.42149999737739563}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8331999778747559},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7533000111579895},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.54339998960495},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5375999808311462},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.5293999910354614},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.517799973487854},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.5041999816894531},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4415999948978424},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.4410000145435333},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4246000051498413},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.421999990940094},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.42149999737739563},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4034000039100647},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.39989998936653137},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.391400009393692},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3806999921798706},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3718000054359436},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.35359999537467957},{"id":"https://openalex.org/C157742956","wikidata":"https://www.wikidata.org/wiki/Q3237776","display_name":"Frequency scaling","level":3,"score":0.3418999910354614},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.32829999923706055},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.32679998874664307},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.30219998955726624},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.28760001063346863},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.28279998898506165},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.28119999170303345},{"id":"https://openalex.org/C2775941552","wikidata":"https://www.wikidata.org/wiki/Q25212305","display_name":"Isolation (microbiology)","level":2,"score":0.27970001101493835},{"id":"https://openalex.org/C98025372","wikidata":"https://www.wikidata.org/wiki/Q477538","display_name":"Systems architecture","level":3,"score":0.2754000127315521},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.26100000739097595},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.26019999384880066},{"id":"https://openalex.org/C118021083","wikidata":"https://www.wikidata.org/wiki/Q610398","display_name":"System on a chip","level":2,"score":0.2583000063896179},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.2517000138759613}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3712285.3759777","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3712285.3759777","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 International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3712285.3759777","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3712285.3759777","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 International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W2124509324","https://openalex.org/W2132234208","https://openalex.org/W2441967103","https://openalex.org/W2963469388","https://openalex.org/W2979602826","https://openalex.org/W2982008795","https://openalex.org/W2992678377","https://openalex.org/W2998702515","https://openalex.org/W3007299504","https://openalex.org/W3028864969","https://openalex.org/W3037277842","https://openalex.org/W3157531038","https://openalex.org/W4214821564","https://openalex.org/W4280568654","https://openalex.org/W4285121610","https://openalex.org/W4306317682","https://openalex.org/W4308083744","https://openalex.org/W4360831829","https://openalex.org/W4380624072","https://openalex.org/W4381329139","https://openalex.org/W4386568791","https://openalex.org/W4388031336","https://openalex.org/W4391971774","https://openalex.org/W4393407040","https://openalex.org/W4395020674","https://openalex.org/W4399174482","https://openalex.org/W4399911364","https://openalex.org/W4400680802","https://openalex.org/W4400909627","https://openalex.org/W4400909968","https://openalex.org/W4401212085","https://openalex.org/W4401568811","https://openalex.org/W4403334538","https://openalex.org/W4407448732"],"related_works":[],"abstract_inverted_index":{"Approximate":[0],"Nearest":[1],"Neighbor":[2],"Search":[3],"(ANNS)":[4],"is":[5],"a":[6,60],"critical":[7,30],"component":[8],"of":[9],"modern":[10],"AI":[11],"systems,":[12],"such":[13],"as":[14],"recommendation":[15],"engines":[16],"and":[17,50,103,153],"retrieval-augmented":[18],"large":[19],"language":[20],"models":[21],"(RAG-LLMs).":[22],"However,":[23],"scaling":[24],"ANNS":[25],"to":[26,47,67,81,99],"billion-entry":[27],"datasets":[28],"exposes":[29],"inefficiencies:":[31],"CPU-based":[32,125],"solutions":[33],"are":[34],"bottlenecked":[35],"by":[36],"memory":[37],"bandwidth":[38],"limitations,":[39],"while":[40,127],"GPU":[41,129],"implementations":[42],"underutilize":[43],"hardware":[44,116],"resources,":[45],"leading":[46],"suboptimal":[48],"performance":[49,130],"energy":[51,134],"consumption.":[52],"To":[53],"address":[54],"these":[55],"challenges,":[56],"we":[57],"introduce":[58],"UpANNS,":[59],"novel":[61],"framework":[62],"leveraging":[63],"Processing-in-Memory":[64],"(PIM)":[65],"architecture":[66],"accelerate":[68],"billion-scale":[69],"ANNS.":[70],"UpANNS":[71,119],"integrates":[72],"four":[73],"key":[74],"innovations,":[75],"including":[76],"1)":[77],"architecture-aware":[78],"data":[79],"placement":[80],"minimize":[82],"latency":[83],"through":[84],"workload":[85],"balancing,":[86],"2)":[87],"dynamic":[88],"resource":[89],"management":[90],"for":[91,108,141,147],"optimal":[92],"PIM":[93],"utilization,":[94],"3)":[95],"co-occurrence":[96],"optimized":[97],"encoding":[98],"reduce":[100],"redundant":[101],"computations,":[102],"4)":[104],"an":[105],"early-pruning":[106],"strategy":[107],"efficient":[109],"top-k":[110],"selection.":[111],"Evaluation":[112],"on":[113],"commercial":[114],"UPMEM":[115],"demonstrates":[117],"that":[118],"achieves":[120],"4.3x":[121],"higher":[122],"QPS":[123],"than":[124],"Faiss,":[126],"matching":[128],"with":[131],"2.3x":[132],"greater":[133],"efficiency.":[135],"Its":[136],"near-linear":[137],"scalability":[138],"ensures":[139],"practicality":[140],"growing":[142],"datasets,":[143],"making":[144],"it":[145],"ideal":[146],"applications":[148],"like":[149],"real-time":[150],"LLM":[151],"serving":[152],"large-scale":[154],"retrieval":[155],"systems.":[156]},"counts_by_year":[{"year":2026,"cited_by_count":6}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-12T00:00:00"}