In a recent episode of the OpenObservability Talks podcast, I sat down with Carl Meadows, Chair of the Governing Board of the OpenSearch Software Foundation, and Pallavi Priyadarshini, Technical Steering Committee member and release manager for OpenSearch 3.0. Our conversation explored the May 2025 launch of OpenSearch 3.0 and how it reflects the project’s evolution under Linux Foundation governance.
As Carl and Pallavi remind the audience, the project joined the Linux Foundation in September 2024, establishing the OpenSearch Software Foundation and strengthening the project’s commitment to transparency and collaboration. Since, a Technical Steering Committee (TSC), made up of representatives from 11 organizations, took the lead on technical direction in advance of the release of OpenSearch 3.0.
The conversation unpacks key drivers of the 3.0 release, including Apache Lucene 10, which brought major search performance improvements. The move to JDK 21 also introduced modern Java features and better performance. These changes warranted a major version bump to preserve semantic versioning principles.
We dive into key architectural upgrades in 3.0 such as reader-writer separation and modernized data transport with gRPC and Protobuf, with contributions driven by Uber, Atlassian, and ByteDance, as well as GPU acceleration for 9.3 faster vector indexing and support for Model Context Protocol (MCP) for seamless, context-aware interactions with language models.
True to the podcast’s theme, we dive into observability advancements across the project, including OpenSearch Dashboards and Piped Processing Language (PPL) — as well as recent developments in Data Prepper, OpenSearch’s telemetry-focused ingestion pipeline.
To learn the latest about the project’s 3.x line, including insight into upcoming priorities for development, give the full episode a watch on YouTube or listen on your favorite podcast platform.
