Data Governance Analyzer for Databricks

Evaluates Unity Catalog governance best practices across 25 checks in 6 categories.

Setup

1. Import this project to your Databricks workspace
2. Run the governance_analysis.ipynb notebook with optional parameters:
   • catalog_name (default: main) - Target catalog for results table
   • schema_name (default: default) - Target schema for results table

Usage

Interactive Mode

When running the notebook interactively, widgets will appear at the top:

• Set catalog_name to your target catalog (default: main)
• Set schema_name to your target schema (default: default)

Programmatic/Job Mode

Pass parameters when running as a job:

dbutils.notebook.run(
    "governance_analysis",
    timeout_seconds=600,
    arguments={"catalog_name": "my_catalog", "schema_name": "my_schema"}
)

Results

• Results saved to {catalog}.{schema}.governance_results Delta table
• Summary by category with pass rates and scores
• Overall governance score calculation

Dashboard

The notebook automatically deploys a Lakeview Dashboard to /Shared/Governance/ with:

• Total Checks Counter: Number of governance checks performed
• Total Score Counter: Aggregated governance score
• Pass Rate Counter: Percentage of checks that passed
• Average Score by Category: Bar chart showing performance across categories
• Status Distribution: Pie chart showing pass/fail distribution
• Detailed Results Table: Full governance check results

Manual Import

If automatic dashboard creation fails, you can manually import:

1. Go to SQL Workspace → Dashboards → Import Dashboard
2. Upload dashboard_template.lvdash.json
3. Update the dataset query to your table: {catalog}.{schema}.governance_results

Categories

• Metastore setup (2 checks)
• Identity (7 checks)
• Managed Storage (5 checks)
• Compute/Cluster Policy (1 check)
• Migration Completeness (3 checks)
• Audit & Lineage Coverage (3 checks)
• Privileges (3 checks)

Files

• governance_analyzer.py: Core governance check functions and dashboard deployment
• governance_analysis.ipynb: Main notebook to run checks and deploy dashboard
• dashboard_template.lvdash.json: Lakeview dashboard definition
• README.md: This documentation