Foundations of Semantic Models

Data Ingestion and Semantic Models with Microsoft Fabric

Alex Kuntz

Head of Cloud Curriculum, DataCamp

Default Semantic Model

Power BI Default Semantic Model

  • Creation: Automatically generated when a Lakehouse or Warehouse is created.
  • Editing: Edited via the SQL Analytics Endpoint - model section.
  • Features:
    • Supports basic relationships and calculations
    • Lacks support for advanced calculations and security
Data Ingestion and Semantic Models with Microsoft Fabric

Custom Semantic Model

Power BI Custom Semantic Model

  • Creation: Created using New Semantic Model option in Lakehouse or Warehouse.
  • Editing: Provides a full-featured Data Model editor.
  • Features:
    • Full control over complex relationships and advanced calculations
    • Supports Advanced Features - calculated columns, hierachies, RLS, etc.
Data Ingestion and Semantic Models with Microsoft Fabric

Brief Overview of Dataset Modes

Import Mode

  • loads data in model; fast queries response.
  • Requires updates to keep data up-to-date.

DirectQuery Mode

  • Queries live data directly from the source.
  • Query Response is slower.

Direct Lake Mode

  • Directly queries data from the OneLake.
  • Real-time access to data with no latency.

Import Mode

DirectQuery Mode

Direct Lake Mode Mode

Data Ingestion and Semantic Models with Microsoft Fabric

Composite Dataset Mode

  • Combines Import and DirectQuery modes in one dataset. Composite Mode

  • Choosing the Mode:

    • Use Import mode for static data and DirectQuery for real-time updates.

  • Key Considerations:

    • Ideal for datasets with mixed data needs.
    • Managing both modes adds complexity.
Data Ingestion and Semantic Models with Microsoft Fabric

User Defined Aggregations

What are Aggregations?

  • Summarizes large data into smaller, easier-to-query tables.

When to Use Aggregations?

  • For large datasets slowing down reports.
  • Build reports on summary-level data (e.g., monthly sales).

Benefits of Aggregations

  • Improves performance by reducing query time and data processing load

User Defined Aggregations

Data Ingestion and Semantic Models with Microsoft Fabric

Large Datasets in Power BI Premium

Large Dataset Format

  • Supports datasets over 10 GB with dedicated storage.
  • Provides highly compressed in-memory storage for faster queries.

Features

  • Allows incremental refresh for growing datasets.
  • Supports XMLA endpoint for efficient write operations.
Data Ingestion and Semantic Models with Microsoft Fabric

Let's practice!

Data Ingestion and Semantic Models with Microsoft Fabric

Preparing Video For Download...