[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81643-en":3,"doc-seo-81643-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},81643,8796095360427,"Lucas Martin","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","A GPU-Accelerated Framework for Multi-Attribute Range Filtered Approximate Nearest Neighbor Search","Range-filtered approximate nearest neighbor search (RFANNS) is vital for modern vector databases but current approaches face index inflation and costly construction. Many systems also depend on CPUs for heavy indexing and query processing, limiting throughput because of restricted memory bandwidth and parallelism. Garfield proposes a GPU-accelerated framework for multi-attribute RFANNS using a lightweight GMG index, efficient GPU traversal, and a cell-oriented out-of-core pipeline for datasets beyond GPU memory. Experiments show 4.4× smaller indexes and 119.8× higher throughput versus state-of-the-art RFANNS.","A GPU-Accelerated Framework for Multi-Attribute Range Filtered Approximate Nearest Neighbor Search  \nZhonggen Li  \nZhejiang University [zgli@zju.edu.cn](zgli@zju.edu.cn)  \nHaoran Yu  \nZhejiang University [hryu@zju.edu.cn](hryu@zju.edu.cn)  \nZixuan Xu Zhejiang University [xuzixuan@zju.edu.cn](xuzixuan@zju.edu.cn)  \nYifan Zhu  \nZhejiang University [xtf_z@zju.edu.cn](xtf_z@zju.edu.cn)  \nYunjun Gao Zhejiang University [gaoyj@zju.edu.cn](gaoyj@zju.edu.cn)  \narXiv :2604 .20121v2 [ cs .DB] 25 Apr 2026  \nAbstract  \nRange-filtered approximate nearest neighbor search (RFANNS) is increasingly critical for modern vector databases. However, existing solutions suffer from severe index inflation and construction overhead. Furthermore, they rely exclusively on CPUs for the heavy indexing and query processing, significantly restricting the throughput due to the limited memory bandwidth and parallelism.  \nIn this paper, we present Garfield, a GPU-accelerated framework for multi-attribute range filtered ANNS that overcomes these bottlenecks through designing a lightweight index structure and hardwareaware execution pipeline. Garfield introduces the GMG index, which partitions data into cells and builds local graph indexes. It guarantees linear storage and indexing overhead by adding a constant number of cross-cell edges. For queries, Garfield utilizes a cluster-guided ordering strategy that reorders query-relevant cells, enabling a highly efficient cell-by-cell traversal on the GPU that aggressively reuses candidates as entry points across cells. To handle datasets exceeding GPU memory, Garfield features a cell-oriented out-of-core pipeline. It dynamically schedules cells to minimize the number of active queries per batch and overlaps GPU computation with CPU-to-GPU index streaming. Extensive evaluations demonstrate that Garfield reduces index size by 4.4×, while delivering 119.8× higher throughput than state-of-the-art RFANNS methods.  \nPVLDB Reference Format:  \nZhonggen Li, Haoran Yu, Zixuan Xu, Yifan Zhu, and Yunjun Gao. A GPUAccelerated Framework for Multi-Attribute Range Filtered Approximate Nearest Neighbor Search. PVLDB, 20(2): XXX-XXX, 2027. doi:XX.XX/XXX.XX  \nPVLDB Artifact Availability:  \nThe source code, data, and/or other artifacts have been made available at [https://github.com/ZJU-DAILY/Garfield](https://github.com/ZJU-DAILY/Garfield).  \n1 Introduction  \nThe rapid growth of unstructured data has made Approximate Nearest Neighbor Search (ANNS) a core component of modern database systems, enabling applications like information retrieval [5, 27, 49], recommendation [18, 35, 53], and retrieval-augmentation generation (RAG) [19, 26, 64]. In real-world scenarios, unstructured objects  \nThis work is licensed under the Creative Commons BY-NC-ND 4.0 International License. Visit [https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[ ](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[this license. For any use beyond those covered by this license](this license. For any use beyond those covered by this license), [obtain permission by](obtain permission by)[emailing info@vldb.org. Copyright](emailing info@vldb.org. Copyright) is held by the owner/author(s). Publication rights licensed to the VLDB Endowment.  \nProceedings of the VLDB Endowment, Vol. 20, No. 2 ISSN 2150-8097 .  \ndoi:XX.XX/XXX.XX  \n(a) Comparison of index size.  \n(b) Comparison of index construction.  \nFigure 1: Vanilla ANNS index (HNSW[33]) vs. RFANN indexes (iRangeGraph[52] and UNIFY[29]) on SIFT1M.  \n(e.g., videos, documents) are often coupled with structured numeric attributes (e.g., timestamp, duration) . Consequently, modern systems increasingly demand range-filtered ANNS (RFANNS) [44, 48], which answer vector similarity queries while enforcing structured attribute range predicates to retrieve more relevant results [25] . For instance, video platforms such as YouTube en","cbCaipHo1oPQ1qab","https://ap.wps.com/l/cbCaipHo1oPQ1qab","pdf",2602355,1,14,"English","en",105,"# Introduction\n## Range-filtered ANNS and application demand\n## Key algorithmic and system bottlenecks\n## Hardware limitations and GPU challenges\n## Garfield contributions","[{\"question\":\"What problem does Garfield address in range-filtered approximate nearest neighbor search (RFANNS)?\",\"answer\":\"Garfield targets RFANNS methods that suffer from severe index inflation, expensive construction overhead, and CPU-bound indexing/query processing that limits throughput.\"},{\"question\":\"What is the GMG index in Garfield?\",\"answer\":\"GMG partitions data into cells and builds local graph indexes, guaranteeing linear storage and indexing overhead by adding only a constant number of cross-cell edges.\"},{\"question\":\"How does Garfield handle datasets larger than GPU memory?\",\"answer\":\"Garfield uses a cell-oriented out-of-core pipeline that dynamically schedules cells and overlaps GPU computation with CPU-to-GPU index streaming.\"}]",1784175100,35,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"a-gpu-accelerated-framework-for-multi-attribute-range-filtered-approximate-nearest-neighbor-search","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/a-gpu-accelerated-framework-for-multi-attribute-range-filtered-approximate-nearest-neighbor-search/81643/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What problem does Garfield address in range-filtered approximate nearest neighbor search (RFANNS)?","Question",{"text":74,"@type":75},"Garfield targets RFANNS methods that suffer from severe index inflation, expensive construction overhead, and CPU-bound indexing/query processing that limits throughput.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What is the GMG index in Garfield?",{"text":79,"@type":75},"GMG partitions data into cells and builds local graph indexes, guaranteeing linear storage and indexing overhead by adding only a constant number of cross-cell edges.",{"name":81,"@type":72,"acceptedAnswer":82},"How does Garfield handle datasets larger than GPU memory?",{"text":83,"@type":75},"Garfield uses a cell-oriented out-of-core pipeline that dynamically schedules cells and overlaps GPU computation with CPU-to-GPU index streaming.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":105,"slug":137},19,"General","general"]