databricks Databricks Release Notes ·

Databricks Serverless Compute Access Control is Now Generally Available

securitydatabricksgaarchitectdatabricks-lakeflow
feature

Databricks has announced the general availability of serverless compute access control. This feature allows workspace administrators to govern user access to serverless workloads for notebooks, Databricks Connect, jobs, and Spark Declarative Pipelines on Lakeflow. Access is managed through two built-in objects, Default Interactive Compute and Default Automated Compute. Existing workloads are unaffected, as all users default to "Can Use" permissions on both objects.

Features (1)
  • Generally Available Serverless Compute Access Control

    Databricks' serverless compute access control is now generally available, enabling workspace administrators to manage who can run serverless workloads. This control applies to notebooks, Databricks Connect, jobs, and Spark Declarative Pipelines on Lakeflow through dedicated Default Interactive Compute and Default Automated Compute objects. All users retain default 'Can Use' permissions to ensure continuity for existing workloads.

Read the original announcement →

https://docs.databricks.com/aws/en/release-notes/product/2026/august#serverless-compute-access-control-is-now-generally-available

Related releases