Gold Layer - Business-Ready Analytics
The Gold layer contains aggregated, business-ready data optimized for analytics and reporting. In BimlFlex, this maps to the Data Mart integration stage using dimensional modeling patterns.
Gold Layer Characteristics
- Business-Aligned: Organized by business domain or use case
- Aggregated: Pre-calculated metrics and summaries
- Optimized: Designed for query performance
- Governed: Access controls and data quality enforced
BimlFlex Implementation
The Gold layer is implemented using Dimensional Modeling (Star Schema) patterns:
| Component | Purpose | BimlFlex Support |
|---|---|---|
| Fact Tables | Measurements and metrics | Fact loading patterns |
| Dimension Tables | Business context and attributes | Type 1 and Type 2 SCD support |
| Bridge Tables | Multi-valued dimensions | Data Vault Bridge patterns |
| Aggregate Tables | Pre-calculated summaries | Custom modeling |
tip
For detailed Data Mart configuration, see the Delivering Data Marts documentation.
Dimensional Model Patterns
Fact Tables
BimlFlex supports multiple fact loading patterns:
- Transaction Facts: Individual business events
- Periodic Snapshot Facts: Point-in-time measurements
- Accumulating Snapshot Facts: Lifecycle tracking
Dimension Tables
Slowly Changing Dimension (SCD) support:
- Type 1: Overwrite (no history)
- Type 2: Add row (full history)
- Type 3: Add column (limited history)
Platform-Specific Considerations
Microsoft Fabric
- Gold tables optimized for Direct Lake mode
- Semantic models (Power BI datasets) built on Gold layer
- Consider Fabric Warehouse for T-SQL analytics
Databricks
- Delta tables with optimized Z-ordering
- SQL Warehouse for BI tool connectivity
- Photon acceleration for complex queries
Snowflake
- Automatic clustering for query performance
- Materialized views for complex aggregations
- Search optimization for selective queries
Best Practices
- Design for consumers: Understand reporting and analytics requirements
- Minimize joins: Denormalize where appropriate for performance
- Pre-aggregate: Calculate common metrics at appropriate grain
- Document business logic: Clear definitions for all measures
- Monitor usage: Optimize based on actual query patterns