Power BI and SQL Data Warehouse allow you to quickly and easily create data visualizations, even at terabyte scale, or contains both relational and non-relational aspects.
Instead of moving data for later analysis, Power BI leverages a technology called Direct Connect which simplifies the process by translating all the actions you complete in Power BI into targeted TSQL. This code is then executed directly against SQL Data Warehouse, taking advantage of its compute power and unparalleled architecture. Therefore, when connecting to SQL Data Warehouse Power BI you will only move the most minimal amount of the data needed to create a visualization. This also means almost all of the processing is completed within SQL Data Warehouse, taking advantage of its parallelized architecture and compute resources.
Getting started with Power BI is easy. Visit PowerBI.com, or click the Open in Power BI button on the main page of your SQL Data Warehouse in the Azure Portal. Once you are connected to Power BI you can use it to:
- Visualize and analyze all of your data
- Interact with your data and see results in real-time
- Create, save, and share visualizations across your organization
If you want to test out connecting to SQL Data Warehouse through Direct Connect, follow the tutorial below or check out our documentation.
PolyBase, Microsoft’s tool for querying non-relational data using SQL, can be used to further enhance the relationship between SQL Data Warehouse and Power BI. With PolyBase, it’s simple to define an External Table in SQL Data Warehouse that references data in your Azure Storage accounts. These external tables will show up alongside your relational data in Power BI and can be analyzed there. Simply follow the tutorial below to get started with PolyBase.
You can even analyze your non-relational and relational data together. Simply define a view or create a table joining a relational table with an external table in order to analyze all your data together, at once. Using PolyBase and Power BI makes it easy to bring this disparate data together and analyze it all at once.