https://opensearch.org/blog/opensearch-3-0-enhances-vector-database-performance/ Skip to main content OpenSearch 3.0 Released * Read Announcement [Start Typing... ] Close Search OpenSearch search Menu * Platform + Platform o [svg]OpenSearchOpenSearch is a powerful search and analytics engine built on Apache Lucene. o [svg]OpenSearch DashboardsOur data visualization toolset is a flexible, fully integrated solution for visually exploring and querying your data. o [svg]OpenSearch Data PrepperA server-side data collector designed to enrich, transform, and aggregate data for downstream analytics with OpenSearch. o Vector Search IconOpenSearch Vector EnginePower machine learning and generative AI applications with a high-performance database designed for the way AI understands data. + Capabilities o [svg]Machine Learning and AI # Vector Search # Anomaly Detection o [svg]Search # E-Commerce # Document Search o [svg]Observability # Performance Monitoring # Log Analysis o [svg]Security Analytics # Threat Intelligence # Event Correlation + Performance Benchmarks o [svg]View key performance metrics across different workloads o View Data * Community + OpenSearch Community o [svg]ForumFind answers to your questions, help others in the community, and join the conversation. o [svg]EventsCommunity Meetings, Development Backlog & Triage, in-person, and virtual events. o [svg]MembersCommunity member profiles. o [svg]Solutions ProvidersFind open-source providers offering solutions and services. o [svg]ProjectsHighlights of projects built by the community. o [svg]User GroupsJoin the OpenSearch Project Meetup Network + Community Resources o [svg]SlackSpeak with other developers. o [svg]GitHub Project OrganizationJoin us for in-person and virtual events to learn the latest about the project. + OpenSearchCon o Image o OpenSearchCon Europe 202530 April - 1 May 2025 Amsterdam, Netherlands * Documentation + Documentation Library o [svg]OpenSearch and DashboardsBuild your OpenSearch solution using core tooling and visualizations. o [svg]OpenSearch Data PrepperFilter, mutate, and sample your data for ingestion into OpenSearch. + Documentation Library II o [svg]OpenSearch BenchmarkMeasure performance metrics for your OpenSearch cluster. o [svg]ClientsInteract with OpenSearch from your application using language APIs. + Roadmap o Image o OpenSearch Project RoadmapRead our blog on 2024-2025 project development plan and beyond! * Blog * Downloads * search Press OpenSearch 3.0 Enhances Vector Database Performance, Search Infrastructure and Scalability to Meet AI-driven Demand By OpenSearchMay 6, 2025No Comments Latest iteration bolsters open, scalable, community-driven search and analytics, enabling sustainable innovation SAN FRANCISCO - May 6, 2025 - The OpenSearch Software Foundation, the vendor-neutral home for the OpenSearch Project, today announced the general availability of OpenSearch 3.0. This major release delivers a 9.5x performance improvement over OpenSearch 1.3, building on benchmarking data that showed earlier iterations of OpenSearch operating 1.6x faster than its closest industry competitor. Today's AI applications -- like generative AI, hybrid search, retrieval-augmented generation (RAG) and recommendation engines -- rely heavily on vector databases to find patterns in massive, complex datasets, but as the number of vectors explodes into the billions, many organizations struggle with speed, spend and scalability. Forrester emphasizes that traditional databases are no longer able to meet the growing demands of generative AI due to limitations in supporting modern vector multidimensional data and performing similarity searches. OpenSearch 3.0 addresses this challenge and enables users to increase efficiency, deliver superior performance, and accelerate AI application development via new data management, AI agent, and vector search capabilities. Simultaneously, enhancements such as GPU-supported capabilities can reduce costs by 3.75x. "The enterprise search market is skyrocketing in tandem with the acceleration of AI, and it is projected to reach $8.9 billion by 2030," said Carl Meadows, Governing Board Chair at the OpenSearch Software Foundation and Director of Product Management at Amazon Web Services (AWS). "OpenSearch 3.0 is a powerful step forward in our mission to support the community with an open, scalable platform built for the future of search and analytics, and it reflects our commitment to open collaboration and innovation that drives real-world impact." Vector engine innovations increase processing speed and efficiency To support its large-scale search platform and manage a vast amount of vector data, OpenSearch introduced GPU-based acceleration, leveraging NVIDIA cuVS for indexing workflows. New vector engine features include: * GPU Acceleration for OpenSearch Vector Engine: Delivers superior performance for large-scale vector workloads while significantly lowering operational spend by reducing index building time. By enabling GPU deployment, this experimental feature heightens performance for data-intensive workloads and accelerates index builds by up to 9.3x. * Model Context Protocol (MCP) Support: Native MCP support allows AI agents to easily communicate with OpenSearch, enabling more comprehensive and customizable AI-powered solutions. * Derived Source: Reduces storage consumption by one-third by removing redundant vector data sources and utilizing primary data to recreate source documents as needed for reindexing or source call back. Data management features optimize resources, enhance flexibility and drive scalability OpenSearch 3.0 provides major advancements in how the platform ingests, transports and manages data including: * Support for gRPC: Enables faster and more efficient data transport and data processing for OpenSearch deployments. This experimental feature provides a new approach to data transport between clients, servers, and node-to-node communications in OpenSearch. * Pull-based Ingestion: Enhances ingestion efficiency and gives OpenSearch more control over the flow of data and when it's retrieved by decoupling data sources and data consumers. This experimental feature also allows users to pull data from streaming systems like Apache Kafka and Amazon Kinesis. * Reader and Writer Separation: Ensures consistent, high-quality performance for indexing and search workloads by configuring each in isolation, allowing both workloads to work at optimal speed and scale, rather than decreasing in efficiency when the other is taxed. * Apache Calcite Integration: Enables intuitive, iterative query building and exploration by integrating the query builder into OpenSearch SQL and PPL. Simplifies use cases for security, observability and log analysis. * Index Type Detection: Enhances productivity by automatically determining whether an OpenSearch index contains log-related data and speeding up log analysis feature selection. Core upgrades help future-proof OpenSearch's search platform and analytics suite Enhancements to the platform's search infrastructure - removing legacy code, adopting a modular architecture and aligning with the latest Java advancements - boosts maintainability, performance potential, and efficiency. Updates include: * Lucene 10 Upgrade: Modernizes the platform's search infrastructure to ensure long-term innovation, improve indexing and search capabilities, and increase performance of parallel task execution. * Java 21 Minimum Supported Runtime: Enables access to modern language features and performance improvements. * Java Platform Module System Support: Improves organization, eliminates top level split packages and creates a foundation for refactoring the monolithic server module into separable libraries. OpenSearch 3.0 is now available. See the official release blog for more information and full release notes. To learn more about the OpenSearch Software Foundation, including how to get involved, become a member or contribute, please visit foundation.opensearch.org/. About the OpenSearch Software Foundation The OpenSearch Software Foundation is a vendor-neutral community for search, analytics, observability, and vector database software. Hosted by the Linux Foundation and supported by premier members such as AWS, SAP and Uber, the OpenSearch Software Foundation works with community maintainers, developers, and member organizations to drive the continued growth of the OpenSearch project. With more than 900 million software downloads since its inception and participation from thousands of contributors, the OpenSearch project and its community are transforming how information is managed and discovered. To learn more, please visit foundation.opensearch.org. The Linux Foundation has registered trademarks and uses trademarks. For a list of trademarks of The Linux Foundation, please see our trademark usage page. Linux is a registered trademark of Linus Torvalds. Media Contact Kristi Piechnik The Linux Foundation kpiechnik@linuxfoundation.org Author * OpenSearch OpenSearch View all posts Share Subscribe for updates, event info, and the latest community news [opensearch_lo] OpenSearch is a community-driven, Apache 2.0-licensed open source search and analytics suite that makes it easy to ingest, search, visualize, and analyze data. * x-twitter * linkedin * youtube * github * slack * mastodon Participate Code of Conduct Forum Project Repo Community Repo Meetup Connect Providers Become a Provider Find a Provider Resources About FAQ Release Schedule Maintenance Policy Documentation Getting Started Testimonials Trademark & Brand Privacy (c) Copyright The Linux Foundation. The OpenSearch Project is a project of The Linux Foundation. For web site terms of use, trademark policy and other policies applicable to The OpenSearch Project please see Linux Foundation Policies. The OpenSearch Project supports the OpenSearch open source project, which has been established as OpenSearch Project a Series of LF Projects, LLC. For policies applicable to the OpenSearch Project a Series of LF Projects, LLC, please see LF Projects, LLC Policies, Privacy Policy, Terms of Use and the OpenSearch Trademark Policy. Close Menu * Platform + Platform o [OpenSearch-Core]OpenSearchOpenSearch is a powerful search and analytics engine built on Apache Lucene. o [OpenSearch-Dashboards]OpenSearch DashboardsOur data visualization toolset is a flexible, fully integrated solution for visually exploring and querying your data. o [OpenSearch-DataPrepper]OpenSearch Data PrepperA server-side data collector designed to enrich, transform, and aggregate data for downstream analytics with OpenSearch. o Vector Search IconOpenSearch Vector EnginePower machine learning and generative AI applications with a high-performance database designed for the way AI understands data. + Capabilities o [OpenSearch-AI-1]Machine Learning and AI # Vector Search # Anomaly Detection o [OpenSearch-Search]Search # E-Commerce # Document Search o [OpenSearch-Observability]Observability # Performance Monitoring # Log Analysis o [OpenSearch-Security]Security Analytics # Threat Intelligence # Event Correlation + Performance Benchmarks o [OpenSearch]View key performance metrics across different workloads o View Data * Community + OpenSearch Community o [OpenSearch]ForumFind answers to your questions, help others in the community, and join the conversation. o [OpenSearch-Events]EventsCommunity Meetings, Development Backlog & Triage, in-person, and virtual events. o [OpenSearch]MembersCommunity member profiles. o [OpenSearch]Solutions ProvidersFind open-source providers offering solutions and services. o [OpenSearch]ProjectsHighlights of projects built by the community. o [OpenSearch]User GroupsJoin the OpenSearch Project Meetup Network + Community Resources o [OpenSearch-Slack]SlackSpeak with other developers. o [OpenSearch-Github]GitHub Project OrganizationJoin us for in-person and virtual events to learn the latest about the project. + OpenSearchCon o Image o OpenSearchCon Europe 202530 April - 1 May 2025 Amsterdam, Netherlands * Documentation + Documentation Library o [OpenSearch-Dashboards-Square]OpenSearch and Dashboards Build your OpenSearch solution using core tooling and visualizations. o [OpenSearch-DataPrepper-1]OpenSearch Data PrepperFilter, mutate, and sample your data for ingestion into OpenSearch. + Documentation Library II o [OpenSearch]OpenSearch BenchmarkMeasure performance metrics for your OpenSearch cluster. o [OpenSearch]ClientsInteract with OpenSearch from your application using language APIs. + Roadmap o Image o OpenSearch Project RoadmapRead our blog on 2024-2025 project development plan and beyond! * Blog * Downloads