From manual permissions to automatic control by tags
When data grows in Databricks, managing permissions on an object-by-object basis is no longer viable.
Industry Steel Multinational
-80% Manual management
↓ -80% Manual permission management
↑ Automatic Tag deployment
↓ -45% Inconsistencies between DEV and PRO
↑ 50% Speed to analyze lineage by use case.
BEFORE
Tags were not automatically propagated between development and production.
Permits had to be applied and reviewed manually.
It was difficult to know which tables were involved in each use case.
NOW
Tags are automatically deployed between environments.
Permissions are applied according to tags and metadata.
Lineage allows filtering tables by use case.
THE PROBLEM
Manually managing permissions does not scale.
Every change required technical intervention.
Permits depended on manual reviews.
Tags did not automatically travel between environments.
THE SOLUTION
Dynamic permissions based on tags.
Automatic tag deployment in Databricks.
Metadata table to relate group, permission and tag.
Periodic script that applies permissions according to the defined configuration.
Filter by tags in the lineage report.
THE IMPACT
More control. Less operational burden.
The organization can manage permissions in a scalable, traceable and automated way, reducing manual work and improving control over who accesses what data.
