Catalog Federation and More: The Apache Polaris™ (Incubating) 1.1 Release

Note: Apache Polaris is currently undergoing incubation at the Apache Software Foundation.

On behalf of the Apache Polaris™ (incubating) community, we’re excited to share that the project has officially released version 1.1.0-incubating (Polaris 1.1)!

If it hasn’t been clear before now, the momentum around the Polaris project is increasing, with version 1.1 coming just a couple short months after the community’s first official 1.0 release in July 2025. (In an exciting development, the community has agreed to rally around a new, accelerated cadence, with new releases anticipated on a monthly basis. From now on, users can expect more frequent updates and new features from the project.)

Version 1.1 brings some much anticipated functionality to the project. So without further ado, let’s take a look at a few of the highlights from this latest release.

Catalog federation

One of the biggest pieces of functionality that the Polaris community has worked toward for this release is around catalog federation. In catalog federation, a single Polaris catalog instance serves as a main, central catalog with other catalogs operating under it. The idea is that Polaris can potentially serve as a single point of contact to bring together existing catalog instances.

In version 1.1, quite a few new features were added in the realm of catalog federation.

• Hive Metastore support: Users can federate to HiveCatalog using the Iceberg REST library. Similar to what is described above, a primary Polaris instance interfaces with a Polaris instance that wraps Hive Metastore. In this release, Polaris supports federating only to a single Hive instance with implicit authentication.

• Implicit authentication type: Implicit is a new authentication type added in this release to support authenticating with federated catalogs without Polaris needing to store these credentials directly. Instead, credentials can be passed to Polaris through environmental variables or configuration files.

• Modularized federation architecture: With a new modularized architecture Polaris is more extensible and now supports multiple catalog types. As part of this, instances of non-REST catalogs are only loaded when initialized and accessed. In the future, once a non-REST catalog is initialized, the calls to both internal and federated catalogs are the same.

• Updated Polaris CLI: And, finally, the Polaris CLI has been updated to include more support for configuring and managing federated catalogs.

New features and improved functionality

Beyond federation support, there’s a lot more to be excited about in version 1.1.

• External identity providers: The Polaris community built on the work that was done in the previous release and added more support for external identity providers. Supported providers now include Google, Okta, Keycloak and Helm chat configuration options, in addition to others. See the documentation for more details.

• Polaris Python client: With the official Polaris Python package, it’s easier than ever for users to install and use Polaris in Python. As a bonus, users will find enhanced error messages in this latest version.

• Expanded S3 compatibility: Polaris now supports the s3a scheme, allowing for seamless integration with Hadoop and data lake ecosystems.

• MinIO support: Version 1.1 also brings brand new support for MinIO. Now users can take advantage of their own private cloud or on-prem environment. The release also includes end-to-end documentation on how to get started.

• Seamless storage credential refresh: Polaris now includes a refresh credentials endpoint in LoadTableResponse for AWS, Azure and GCP, enabling automatic per-table credential renewal on the client side. This ensures that long-running jobs can seamlessly refresh their storage credentials without interruption, improving both reliability and security.

Get involved

New features are being implemented for Polaris at an exciting pace. We encourage you to download the latest version and give Polaris a try. Be sure to monitor the Github repository for new features and fixes, and join the discussion — because the best time to get involved is now!