What’s New in Apache Polaris 1.4: Advanced Multi-Tenancy and Streamlined Deployment

What to expect in Polaris 1.4

Here is what you can expect in the latest version.

Advanced security and multi-tenancy

Version 1.4 introduces granular control over how Polaris interacts with cloud storage, enabling multi-tenancy with support for separate encryption keys and IAM identities across catalogs.

• Storage-scoped AWS credentials: Users will find new advanced security capabilities for interacting with cloud storage.
• AWS STS session tags: Users can now assign specific fields such as realm, catalog and table for AWS STS session tags.
• Sub-scoped credentials: Principal names can now be used in sub-scoped credentials to improve auditing and reduce the likelihood that cached credentials are shared between different principals.
• Enhanced KMS support: Version 1.4 adds support for S3 data encryption using KMS properties in the catalog storage configuration, complemented by new KMS CLI support for direct management.
• Azure ADLS hierarchical flag: For Azure users, a new hierarchical flag for ADLS storage allows for more granular SAS token down-scoping to align with Azure Data Lake structures.

Persisting query metrics data

Building on the metrics capabilities of previous versions, Polaris 1.4 enhances how query data insights are handled. Users now have the ability to persist Iceberg metrics such as ScanReports and CommitReports directly to the Polaris database. Metrics reporting specifically checks that users have the TABLE_READ_DATA or TABLE_WRITE_DATA privileges. As you test our persistent metrics, note that the feature is in beta.

Streamlined deployment and management

In 1.4, the deployment process for Polaris has been further refined and simplified.

• Enhanced Helm charts: The Helm chart now includes a JSON schema file to prevent misconfiguration and adds native support for the modern Kubernetes Gateway API.
• Python package management: Polaris has switched from Poetry to UV to make deployments faster and easier.
• Postgres optimization: Users will find new performance optimizations on heavy workloads on Postgres.
• Expanded backend database support: The community has launched support for CockroachDB as a backend database. Using MongoDB as a Polaris backend is currently in beta.

Breaking changes and deprecations

To plan your upgrade to 1.4 effectively, please be aware of the following changes:

• Rate-limiter update: The custom token-bucket-based rate limiter has been replaced by Guava's rate limiter. The configuration option polaris.rate-limiter.token-bucket.window is no longer supported.
• Deprecations: PolarisConfigurationStore is being phased out and is now officially marked as deprecated.
• GCP configuration: The gcpServiceAccount configuration value is now active and enables service account impersonation; previously, this value was defined but not used.
• Helm strict types: Existing values files for Helm may require updates to match the new strict types defined in the JSON schema.

Looking ahead

Polaris 1.4 sets a high bar for 2026, focusing heavily on features that matter for production-grade environments. We invite you to download the new version, test out the latest features and join the community.

Apache® and Apache Polaris are either registered trademarks or trademarks of the Apache Software Foundation in the United States and/or other countries.