Apache Polaris™ (Incubating) 1.0 Released: A New Era for the Open Source Catalog

This week marks a major milestone for the open source community: Apache Polaris™ (incubating) has officially released version 1.0.0-incubating (Polaris 1.0), delivering the project’s first production-ready release. Now users can take advantage of a single, downloadable binary artifact to quickly get started with Polaris or leverage the published Helm Chart for drop‐in deployment to any Kubernetes cluster.

Designed to unify data governance and cross-platform intelligence, Polaris is a REST-based catalog service built to serve the needs of modern, open data lakehouses. The 1.0 release is packed with essential features, performance upgrades and powerful new capabilities — many of which were directly shaped by community feedback. 

Whether you’re a developer, an administrator or just Polaris-curious, there’s something in this release for you. Here are just a few of the highlights.

New features and improved functionality

Identity provider support

One of the most requested features is here: support for external identity providers such as Okta and Google. Expanding on the internal, built-in support, Polaris now lets admins configure identity handling at the realm level with both external and hybrid (mixed) modes. This flexibility makes it easier to integrate Polaris into real-world enterprise authentication environments.

Versioned policies and admin endpoints

The new Polaris policy store brings persistent, schema-evolving support for a broad range of policies including data compaction, snapshot expiration and more. Policymakers, administrators and query engines now have a unified set of REST CRUD endpoints to easily manage policy lifecycles. 

With this addition, users can immediately take advantage of governed table maintenance, with support for more policy types such as fine-grained access control (FGAC) coming in the future. Overall, this change enables streamlined data governance use cases such as retention, compaction and operational consistency, which pave the way for scalable, policy-driven data platforms.

Performance enhancements

The 1.0 release introduces a JDBC-based persistence layer, unlocking better concurrency, scalability and reliability. For more details on the improvement, check out this blog.

Additional performance enhancements in this release include:

• HTTP caching with Etags, allowing efficient metadata refreshes via the loadTable endpoint
  

• Conflict resolution improvements, prioritizing user writes over background operations such as compaction
  

• Snapshot filtering, which lets you fetch only the branches or tags you care about — especially useful for complex data workflows by reducing the table metadata size
  

• Migration to the Quarkus runtime, delivering sub-second cold starts, lower memory usage and built-in Kubernetes readiness

Experimental features worth exploring

As part of the 1.0 release, we’re also excited to introduce several beta features that hint at the future of Polaris:

• Generic table support for Delta Lake tables, with an extended REST API and dedicated Spark 3.5 plugin already available and support for CSV file tables and Parquet file tables on the roadmap
  

• An event listener framework for subscribing to catalog-level events such as table commits or view refreshes, enabling extensible downstream workflows
  

• Catalog federation to route requests to external Iceberg REST or Hadoop catalogs, bringing Polaris closer to being your universal catalog layer and providing better support for catalog migration

These features are experimental today but offer a glimpse into where Polaris is heading: toward becoming a central, multiformat, cross-platform open catalog.

Get involved

Apache Polaris 1.0 isn’t just a release — it’s a statement of what’s possible when the open source community comes together. Whether you want to try it out, contribute features, report bugs or just follow along, now is the perfect time to get involved.

Check out the GitHub repo, review the release notes, download it and try it out, and let us know what you think!