{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T19:06:49Z","timestamp":1779131209243,"version":"3.51.4"},"reference-count":61,"publisher":"Association for Computing Machinery (ACM)","issue":"3","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. ACM Manag. Data"],"published-print":{"date-parts":[[2026,5,18]]},"abstract":"<jats:p>Filtered Vector Search (FVS) is critical for supporting semantic search and GenAI applications in modern database systems. However, existing research most often evaluates algorithms in specialized libraries, making optimistic assumptions that do not align with enterprise-grade database systems. Our work challenges this premise by demonstrating that in a production-grade database system, commonly made assumptions do not hold, leading to performance characteristics and algorithmic trade-offs that are fundamentally different from those observed in isolated library settings. This paper presents the first in-depth analysis of filter-agnostic FVS algorithms within a production PostgreSQL-compatible system. We systematically evaluate post-filtering and inline-filtering strategies across a wide range of selectivities and correlations.<\/jats:p>\n                  <jats:p>Our central finding is that the optimal algorithm is not dictated by the cost of distance computations alone, but that system-level overheads that come from both distance computations and filter operations (like page accesses and data retrieval) play a significant role. We demonstrate that graph-based approaches (such as NaviX\/ACORN) can incur prohibitive numbers of filter checks and system-level overheads, compared with clustering-based indexes such as ScaNN, often canceling out their theoretical benefits in real-world database environments.<\/jats:p>\n                  <jats:p>Ultimately, our findings provide the database community with crucial insights and practical guidelines, demonstrating that the optimal choice for a filter-agnostic FVS algorithm is not absolute, but rather a system-aware decision contingent on the interplay between workload characteristics and the underlying costs of data access in a real-world database architecture.<\/jats:p>","DOI":"10.1145\/3802011","type":"journal-article","created":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T18:19:16Z","timestamp":1779128356000},"page":"1-26","source":"Crossref","is-referenced-by-count":0,"title":["An In-Depth Study of Filter-Agnostic Vector Search on a PostgreSQL Database System: [Experiments &amp; Analysis]"],"prefix":"10.1145","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-5901-3079","authenticated-orcid":false,"given":"Duo","family":"Lu","sequence":"first","affiliation":[{"name":"Brown University, Providence, Rhode Island, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2052-8107","authenticated-orcid":false,"given":"Helena","family":"Caminal","sequence":"additional","affiliation":[{"name":"Google, Sunnyvale, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9616-6210","authenticated-orcid":false,"given":"Manos","family":"Chatzakis","sequence":"additional","affiliation":[{"name":"Universit\u00e9 Paris Cit\u00e9, LIPADE, F-75006 Paris, France"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-6360-9496","authenticated-orcid":false,"given":"Yannis","family":"Papakonstantinou","sequence":"additional","affiliation":[{"name":"Google, San Diego, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2214-6919","authenticated-orcid":false,"given":"Yannis","family":"Chronis","sequence":"additional","affiliation":[{"name":"ETH Zurich, Zurich, Switzerland and Google, Zurich, Switzerland"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-8069-5928","authenticated-orcid":false,"given":"Vaibhav","family":"Jain","sequence":"additional","affiliation":[{"name":"Google, Bangalore, India"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4418-4724","authenticated-orcid":false,"given":"Fatma","family":"\u00d6zcan","sequence":"additional","affiliation":[{"name":"Google, Sunnyvale, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2026,5,18]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"2022. Cohere Wikipedia Embeddings Dataset. https:\/\/huggingface.co\/datasets\/Cohere\/wikipedia-22-12\/tree\/main\/en."},{"key":"e_1_2_1_2_1","unstructured":"2024. Pinecone vs. Postgres pgvector: For vector search easy isn't so easy. https:\/\/www.pinecone.io\/blog\/pinecone-vs-pgvector\/."},{"key":"e_1_2_1_3_1","unstructured":"2024. wiki-ann. Hugging Face. https:\/\/huggingface.co\/2024annonymous\/wiki-ann"},{"key":"e_1_2_1_4_1","unstructured":"2025. hnswlib. https:\/\/github.com\/nmslib\/hnswlib."},{"key":"e_1_2_1_5_1","unstructured":"2025. kNN search in Elasticsearch. Elastic Docs. https:\/\/www.elastic.co\/docs\/solutions\/search\/vector\/knn"},{"key":"e_1_2_1_6_1","unstructured":"2025. OpenAI5M. https:\/\/huggingface.co\/datasets\/allenai\/c4."},{"key":"e_1_2_1_7_1","unstructured":"2025. pgvector: Open-source vector similarity search for Postgres. GitHub repository. https:\/\/github.com\/pgvector\/pgvector"},{"key":"e_1_2_1_8_1","unstructured":"2025. Run Vector Search Queries (Atlas Vector Search stage). MongoDB Documentation. https:\/\/www.mongodb.com\/docs\/atlas\/atlas-vector-search\/vector-search-stage\/"},{"key":"e_1_2_1_9_1","unstructured":"2025. ScaNN. github.com\/google-research\/google-research\/tree\/master\/scann."},{"key":"e_1_2_1_10_1","unstructured":"2025. ScaNN for AlloyDB. https:\/\/services.google.com\/fh\/files\/misc\/scann_for_alloydb_whitepaper.pdf."},{"key":"e_1_2_1_11_1","unstructured":"2025. Supercharging vector search performance and relevance with pgvector 0.8.0 on Amazon Aurora Post-greSQL. https:\/\/aws.amazon.com\/blogs\/database\/supercharging-vector-search-performance-and-relevance-with- pgvector-0-8-0-on-amazon-aurora-postgresql\/."},{"key":"e_1_2_1_12_1","unstructured":"2025. Vector Search in the Real World: How to Filter Efficiently Without Killing Recall. https:\/\/milvus.io\/blog\/how-to-filter-efficiently-without-killing-recall.md."},{"key":"e_1_2_1_13_1","unstructured":"2025. Weviate. https:\/\/weaviate.io\/blog\/speed-up-filtered-vector-search."},{"key":"e_1_2_1_14_1","volume-title":"Gemini: A Large Language Model. https:\/\/geminilang.google","author":"Google","year":"2023","unstructured":"Google AI. 2023. Gemini: A Large Language Model. https:\/\/geminilang.google"},{"key":"e_1_2_1_15_1","volume-title":"Proc. ACM Manag. Data","author":"Aomar Anas Ait","year":"2025","unstructured":"Anas Ait Aomar, Karima Echihabi, Marco Arnaboldi, Ioannis Alagiannis, Damien Hilloulin, and Manal Cherkaoui. 2025. RWalks: Random Walks as Attribute Diffusers for Filtered Vector Search. Proc. ACM Manag. Data (2025)."},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1145\/2783258.2783405"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.is.2021.101807"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3709693"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3698822"},{"key":"e_1_2_1_20_1","volume-title":"Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V. 2. 5299-5310","author":"Ceccarello Matteo","year":"2025","unstructured":"Matteo Ceccarello, Alexandra Levchenko, Ioana Ileana, and Themis Palpanas. 2025. Evaluating and generating query workloads for high dimensional vector similarity search. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V. 2. 5299-5310."},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/3749160"},{"key":"e_1_2_1_22_1","volume-title":"Return of the lernaean hydra: Experimental evaluation of data series approximate similarity search. arXiv preprint arXiv:2006.11459","author":"Echihabi Karima","year":"2020","unstructured":"Karima Echihabi, Kostas Zoumpatianos, Themis Palpanas, and Houda Benbrahim. 2020. Return of the lernaean hydra: Experimental evaluation of data series approximate similarity search. arXiv preprint arXiv:2006.11459 (2020)."},{"key":"e_1_2_1_23_1","volume-title":"Filtered- DiskANN: Graph Algorithms for Approximate Nearest Neighbor Search with Filters (WWW '23)","author":"Gollapudi Siddharth","year":"2023","unstructured":"Siddharth Gollapudi, Neel Karia, Varun Sivashankar, Ravishankar Krishnaswamy, Nikit Begwani, Swapnil Raz, Yiyong Lin, Yin Zhang, Neelam Mahapatro, Premkumar Srinivasan, Amit Singh, and Harsha Vardhan Simhadri. 2023. Filtered- DiskANN: Graph Algorithms for Approximate Nearest Neighbor Search with Filters (WWW '23)."},{"key":"e_1_2_1_24_1","volume-title":"On the difficulty of nearest neighbor search. arXiv preprint arXiv:1206.6411","author":"He Junfeng","year":"2012","unstructured":"Junfeng He, Sanjiv Kumar, and Shih-Fu Chang. 2012. On the difficulty of nearest neighbor search. arXiv preprint arXiv:1206.6411 (2012)."},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3403305"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2011.5946540"},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/3725399"},{"key":"e_1_2_1_28_1","unstructured":"Jonathan Katz. 2024. Scalar and Binary Quantization for pgvector Vector Search and Storage. https:\/\/jkatz.github.io\/ post\/postgres\/pgvector-scalar-binary-quantization\/. Accessed: 2026-01-27."},{"key":"e_1_2_1_29_1","first-page":"9459","article-title":"Retrieval-augmented generation for knowledge-intensive nlp tasks","volume":"33","author":"Lewis Patrick","year":"2020","unstructured":"Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich K\u00fcttler, Mike Lewis, Wen-tau Yih, Tim Rockt\u00e4schel, et al. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in Neural Information Processing Systems 33 (2020), 9459-9474.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_1_30_1","unstructured":"Mocheng Li Xiao Yan Baotong Lu Yue Zhang James Cheng and Chenhao Ma. 2025. Attribute Filtering in Approximate Nearest Neighbor Search: An In-depth Experimental Study. arXiv:2508.16263 [cs.DB]"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2019.2909204"},{"key":"e_1_2_1_32_1","volume-title":"SIEVE: Effective Filtered Vector Search with Collection of Indexes. arXiv preprint arXiv:2507.11907","author":"Li Zhaoheng","year":"2025","unstructured":"Zhaoheng Li, Silu Huang, Wei Ding, Yongjoo Park, and Jianjun Chen. 2025. SIEVE: Effective Filtered Vector Search with Collection of Indexes. arXiv preprint arXiv:2507.11907 (2025)."},{"key":"e_1_2_1_33_1","doi-asserted-by":"crossref","first-page":"1118","DOI":"10.14778\/3717755.3717770","article-title":"UNIFY: Unified Index for Range Filtered Approximate Nearest Neighbors Search","volume":"18","author":"Liang Anqi","year":"2025","unstructured":"Anqi Liang, Pengcheng Zhang, Bin Yao, Zhongpu Chen, Yitong Song, and Guangxu Cheng. 2025. UNIFY: Unified Index for Range Filtered Approximate Nearest Neighbors Search. Proceedings of the VLDB Endowment 18, 4 (2025), 1118-1130.","journal-title":"Proceedings of the VLDB Endowment"},{"key":"e_1_2_1_34_1","volume-title":"Proceedings of the 16th Annual Conference on Innovative Data Systems Research (CIDR '26)","author":"Liu Jiayi","year":"2026","unstructured":"Jiayi Liu, Yunan Zhang, Chenzhe Jin, Aditya Gupta, Shige Liu, and Jianguo Wang. 2026. Fast Vector Search in PostgreSQL: A Decoupled Approach. In Proceedings of the 16th Annual Conference on Innovative Data Systems Research (CIDR '26). Chaminade, USA. https:\/\/www.cidrdb.org\/papers\/2026\/p2-liu.pdf"},{"key":"e_1_2_1_35_1","doi-asserted-by":"crossref","unstructured":"Yu A. Malkov and D. A. Yashunin. 2020. Efficient and Robust Approximate Nearest Neighbor Search Using Hierarchical Navigable Small World Graphs. IEEE Trans. Pattern Anal. Mach. Intell. (2020).","DOI":"10.1109\/TPAMI.2018.2889473"},{"key":"e_1_2_1_36_1","unstructured":"OpenAI. 2024. ChatGPT (November 2024 version). https:\/\/openai.com"},{"key":"e_1_2_1_37_1","unstructured":"Oracle Corporation. 2025. Oracle AI Vector Search. https:\/\/www.oracle.com\/database\/ai-vector-search\/."},{"key":"e_1_2_1_38_1","unstructured":"Oracle Corporation. 2025. Oracle Vector Search Manual. https:\/\/docs.oracle.com\/en\/database\/oracle\/oracle-database\/26\/vecse\/ai-vector-search-users-guide.pdf."},{"key":"e_1_2_1_39_1","unstructured":"Yannis Papakonstantinou Anastasia Ailamaki Yannis Chronis Helena Caminal and Fatma Ozcan. 2025. Filtered Vector Search: State-of-the-art and Research Opportunities."},{"key":"e_1_2_1_40_1","volume-title":"Proc. ACM Manag. Data","author":"Patel Liana","year":"2024","unstructured":"Liana Patel, Peter Kraft, Carlos Guestrin, and Matei Zaharia. 2024. ACORN: Performant and Predicate-Agnostic Search Over Vector Embeddings and Structured Data. Proc. ACM Manag. Data (2024)."},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.14778\/3748191.3748193"},{"key":"e_1_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1162"},{"key":"e_1_2_1_43_1","volume-title":"Pinecone: The Vector Database for AI Search and Retrieval. https:\/\/www.pinecone.io\/.","year":"2025","unstructured":"Pinecone. 2025. Pinecone: The Vector Database for AI Search and Retrieval. https:\/\/www.pinecone.io\/."},{"key":"e_1_2_1_44_1","unstructured":"Abdel Rodriguez John Trengrove and Joon-Pil (JP) Hwang. 2024. How we speed up filtered vector search with ACORN. https:\/\/weaviate.io\/blog\/speed-up-filtered-vector-search. Accessed: 2024-11-19."},{"key":"e_1_2_1_45_1","volume-title":"Proc. VLDB Endow.","author":"Sehgal Gaurav","year":"2025","unstructured":"Gaurav Sehgal and Semih Saliho\u011flu. 2025. NaviX: A Native Vector Index Design for Graph DBMSs With Robust Predicate-Agnostic Search Performance. Proc. VLDB Endow. (2025)."},{"key":"e_1_2_1_46_1","first-page":"177","article-title":"Results of the NeurIPS'21 challenge on billion-scale approximate nearest neighbor search. In NeurIPS 2021 Competitions and Demonstrations Track","author":"Simhadri Harsha Vardhan","year":"2022","unstructured":"Harsha Vardhan Simhadri, George Williams, Martin Aum\u00fcller, Matthijs Douze, Artem Babenko, Dmitry Baranchuk, Qi Chen, Lucas Hosseini, Ravishankar Krishnaswamny, Gopal Srinivasa, et al. 2022. Results of the NeurIPS'21 challenge on billion-scale approximate nearest neighbor search. In NeurIPS 2021 Competitions and Demonstrations Track. PMLR, 177-189.","journal-title":"PMLR"},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.1145\/2812802"},{"key":"e_1_2_1_48_1","volume-title":"Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971","author":"Touvron Hugo","year":"2023","unstructured":"Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timoth\u00e9e Lacroix, Baptiste Rozi\u00e8re, Naman Goyal, Eric Hambro, Faisal Azhar, et al. 2023. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 (2023)."},{"key":"e_1_2_1_49_1","doi-asserted-by":"crossref","unstructured":"Nitish Upreti Harsha Vardhan Simhadri Hari Sudan Sundar Krishnan Sundaram Samer Boshra Balachandar Perumalswamy Shivam Atri Martin Chisholm Revti Raman Singh Greg Yang Tamara Hass Nitesh Dudhey Subramanyam Pattipaka Mark Hildebrand Magdalen Manohar Jack Moffitt Haiyang Xu Naren Datha Suryansh Gupta Ravishankar Krishnaswamy Prashant Gupta Abhishek Sahu Hemeswari Varada Sudhanshu Barthwal Ritika Mor James Codella Shaun Cooper Kevin Pilch Simon Moreno Aayush Kataria Santosh Kulkarni Neil Deshpande Amar Sagare Dinesh Billa Zishan Fu and Vipul Vishal. 2025. Cost-Effective Low Latency Vector Search with Azure Cosmos DB. (2025).","DOI":"10.14778\/3750601.3750635"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1145\/3725325"},{"key":"e_1_2_1_51_1","volume-title":"Milvus: A Purpose-Built Vector Data Management System (SIGMOD '21)","author":"Wang Jianguo","year":"2021","unstructured":"Jianguo Wang, Xiaomeng Yi, Rentong Guo, Hai Jin, Peng Xu, Shengjun Li, Xiangyu Wang, Xiangzhou Guo, Chengming Li, Xiaohai Xu, Kun Yu, Yuxing Yuan, Yinghao Zou, Jiquan Long, Yudong Cai, Zhenxiang Li, Zhifeng Zhang, Yihua Mo, Jun Gu, Ruiyi Jiang, Yi Wei, and Charles Xie. 2021. Milvus: A Purpose-Built Vector Data Management System (SIGMOD '21)."},{"key":"e_1_2_1_52_1","unstructured":"Mengzhao Wang Lingwei Lv Xiaoliang Xu Yuxiang Wang Qiang Yue and Jiongkang Ni. 2023. An efficient and robust framework for approximate nearest neighbor search with attribute constraint (NIPS '23)."},{"key":"e_1_2_1_53_1","first-page":"3","article-title":"Graph-and Tree-based Indexes for High-dimensional Vector Similarity Search: Analyses, Comparisons, and Future Directions","volume":"46","author":"Wang Zeyu","year":"2023","unstructured":"Zeyu Wang, Peng Wang, Themis Palpanas, and Wei Wang. 2023. Graph-and Tree-based Indexes for High-dimensional Vector Similarity Search: Analyses, Comparisons, and Future Directions. IEEE Data Eng. Bull. 46, 3 (2023), 3-21.","journal-title":"IEEE Data Eng. Bull."},{"key":"e_1_2_1_54_1","doi-asserted-by":"publisher","DOI":"10.14778\/3704965.3704974"},{"key":"e_1_2_1_55_1","doi-asserted-by":"publisher","DOI":"10.1145\/3709729"},{"key":"e_1_2_1_56_1","doi-asserted-by":"publisher","DOI":"10.1145\/3698814"},{"key":"e_1_2_1_57_1","doi-asserted-by":"publisher","DOI":"10.1145\/3709679"},{"key":"e_1_2_1_58_1","doi-asserted-by":"crossref","first-page":"1536","DOI":"10.14778\/3718057.3718078","article-title":"BigVectorBench: Heterogeneous Data Embedding and Compound Queries are Essential in Evaluating Vector Databases","volume":"18","author":"Zhan Chaoqun","year":"2025","unstructured":"Chaoqun Zhan, Mengzhao Wang, Lingwei Lv, Yitong Geng, Bin Wu, and Sheng Wang. 2025. BigVectorBench: Heterogeneous Data Embedding and Compound Queries are Essential in Evaluating Vector Databases. Proceedings of the VLDB Endowment 18, 5 (2025), 1536-1549.","journal-title":"Proceedings of the VLDB Endowment"},{"key":"e_1_2_1_59_1","volume-title":"Proceedings of the ACM on Management of Data","author":"Zhang Fangyuan","year":"2025","unstructured":"Fangyuan Zhang, Mengxu Jiang, Guanhao Hou, Jieming Shi, Hua Fan, Wenchao Zhou, Feifei Li, and Sibo Wang. 2025. Efficient Dynamic Indexing for Range Filtered Approximate Nearest Neighbor Search. Proceedings of the ACM on Management of Data (2025)."},{"key":"e_1_2_1_60_1","volume-title":"2024 IEEE 40th International Conference on Data Engineering (ICDE).","author":"Zhang Yunan","year":"2024","unstructured":"Yunan Zhang, Shige Liu, and Jianguo Wang. 2024. Are There Fundamental Limitations in Supporting Vector Data Management in Relational Databases? A Case Study of PostgreSQL. In 2024 IEEE 40th International Conference on Data Engineering (ICDE)."},{"key":"e_1_2_1_61_1","doi-asserted-by":"publisher","DOI":"10.1145\/3639324"}],"container-title":["Proceedings of the ACM on Management of Data"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3802011","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T18:28:52Z","timestamp":1779128932000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3802011"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,18]]},"references-count":61,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,5,18]]}},"alternative-id":["10.1145\/3802011"],"URL":"https:\/\/doi.org\/10.1145\/3802011","relation":{},"ISSN":["2836-6573"],"issn-type":[{"value":"2836-6573","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,18]]}}}