Introduction
Organizations store large volumes of data in warehouses and lakehouses, but raw data rarely arrives in the shape that analysts and downstream systems need. T-SQL is the standard language for querying and transforming relational data. In Microsoft Fabric, you can run T-SQL in both warehouses and lakehouses. A warehouse provides full read-write T-SQL capabilities, while a lakehouse exposes its Delta tables through a read-only SQL analytics endpoint for querying and analysis. This module focuses on transforming data within a Fabric warehouse, where read-write support lets you persist transformation results.
Imagine your organization recently migrated its data into a Fabric warehouse. Staging tables hold raw sales, customer, and product data from multiple source systems. The analytics team needs clean, well-structured data for reporting. Your machine learning team needs consistent, aggregated datasets. Right now, every analyst writes their own ad hoc queries, leading to duplicated logic, inconsistent results, and no repeatable process for refreshing data.
In this module, you explore how to use T-SQL to transform data in a Fabric warehouse. You start with query-based transformations such as filtering, joining, and aggregating data. Then you create views that encapsulate reusable logic, and build stored procedures that automate repeatable data processing tasks. Finally, you implement dimensional tables (fact and dimension tables) that serve as the foundation for semantic models and analytics.
By the end of this module, you're able to transform warehouse data using T-SQL queries, create views and stored procedures for reusable logic, and implement dimensional tables for analytics.