# Performance of OpenSearch K-NN Search with Additional Filters (e.g., Metadata Conditions) **URL:** \ **Category:** k-NN\ **Created:** [July 6, 2025, 6:15pm UTC](https://forum.opensearch.org/t/performance-of-opensearch-k-nn-search-with-additional-filters-e-g-metadata-conditions/25589 "2025-07-06T18:15:26Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![bhuvan](https://avatars.discourse-cdn.com/v4/letter/b/f4b2a3/32.png) [@bhuvan](https://forum.opensearch.org/u/bhuvan)\ **Post date:** [July 6, 2025, 6:15pm UTC](https://forum.opensearch.org/t/performance-of-opensearch-k-nn-search-with-additional-filters-e-g-metadata-conditions/25589/1 "2025-07-06T18:15:26Z")
I came across [this AWS blog](https://aws.amazon.com/blogs/big-data/choose-the-k-nn-algorithm-for-your-billion-scale-use-case-with-opensearch/) on OpenSearch K-NN — very helpful! I’m exploring K-NN queries with filters (e.g., `status = 'active'`) and wondering about performance implications. In pgvector, there’s an **“iterative index scan”** explained in detail in [this blog](https://aws.amazon.com/blogs/database/supercharging-vector-search-performance-and-relevance-with-pgvector-0-8-0-on-amazon-aurora-postgresql/) that optimizes K-NN with filters. **Does OpenSearch offer a similar optimization or guidance for such queries?** Any benchmarks or best practices would be great. ---
**Author:** ![system](https://us1.discourse-cdn.com/flex019/uploads/mauve_hedgehog/original/2X/3/33547ea01a5b12dcca2958411d3edd97ae2ea8c1.png) [@system](https://forum.opensearch.org/u/system)\ **Post date:** [September 4, 2025, 6:15pm UTC](https://forum.opensearch.org/t/performance-of-opensearch-k-nn-search-with-additional-filters-e-g-metadata-conditions/25589/2 "2025-09-04T18:15:41Z")
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.