Power BI Datamart — an overview

Since Microsoft introduced a Datamart we can say that Power BI ecosystem is a full cloud experience. Power BI Desktop in the cloud I can state. How it works? Once you create and open a Datamart you are welcomed by well-known ribbon where you can work with data.

Imported data is prepared and transformed by Power Query.

You can manage Row Level Security options.

Create new measures with a full intellisense.

Because all Datamart data is saved inside a separate Azure SQL Database we can query the data by T-SQL.

If you prefer to use an external application like Azure Data Studio, SSMS, etc. You can obtain a connection address from Datamart settings and connect to the database directly where you have an access to views.

For people who prefer to use Power Query, there is a Visual Query editor.

The last section is a model editor, where we can work with the model.

Every Datamart creates its own dataset.

As you see on the screen, a dataset owner is the Datamart itself. So there is no possibility to change its settings.

You are able to connect to the Dataset from Power BI Desktop. Of course you can connect to the Datamart as well. Both of them are connected as Live Connection storage mode.

You can query the model from DAX Studio as well.

As I mentioned earlier every table in the model is saved inside an Azure SQL database. It is worth mentioning that all the model tables storage mode is Direct Query (no changeable). So you need to consider if this setup fulfills your performance requirements. And general Direct Query limitations.

When it comes to administration I can say that Datamart is an additional layer on top of a dataset, so settings are similar like Query Caching, Scheduled refreshing, etc.