Analysis Services LiveConnection

What is Analysis Services LiveConnection

LiveConnection is a type of connection to a data source. This type of connection does not store a second copy of the data into the memory. Instead, data remains in the data source, and visualizations will continuously query the data source from Power BI.
This is the fastest way to connect Power BI to a source. It also gives extra benefits: your model can be stored in source control, you can use partitioning and scaling on your model, and easily connect tools like Excel to it.

There are 3 flavors:

• Azure Analysis Services – Tabular (AAS).
  This is a PAAS environment in Azure, where you have a pay-per-use model. Power BI Datasets are also based on Tabular models. So the same syntax (DAX) is used to describe measures. You can even load a Power BI Dataset in Analysis Services. All security is managed through Azure Active Directory (AAD).
• SQL Server Analysis Services – Tabular (SSAS-T).
  The big difference with AAS is that it’s an ‘on-premise’ solution, where you have to manage SQL Server and the OS yourself. The security is set through Active Directory. This gives some complications because Power BI works with AAD.
• SQL Server Analysis Services – Dimensional Cude (SSAS-DC).
  The difference here is that the model language is in MDX, whereas the Power BI language is in DAX. Because of this, importing Power BI models is not possible. Plus, you would also have to learn another syntax.

Therefore, the first option, Azure Analysis services, works best with Power BI and Webdashboard. The on-premise variants will need some extra work.

LiveConnection vs Import/DirectQuery

There is a big technical difference between LiveConnection and Import/DirectQuery. With LiveConnection you connect directly to the Model Structure (dataset) on Analysis Services. You also go through Power BI with the same authorization token. The dataset you see in Power BI, isn’t used at runtime, only at design time.
For both Import and DirectQuery, you always only connect to Power BI, because the Model Structure is located in Power BI.

Configuring Azure Analysis Services

Webdashboard in a Standard environment

Webdashboard creates a Publish Account for you which includes a Power BI Pro license. When connecting to your datasets (Model Structure), Webdashboard will use that Publish Account to gain access. This works for Direct Query and Import because the model structure is inside Power BI (see figure on the previous page). In the case of a LiveConnection the Model Structure is located in a different Azure Active Directory. To make this work, 2 steps are needed:

1. Add the Publish Account as a guest to your Azure Active Directory (and accept the invite).
   You can find your Publish Account under Gear icon -> Settings -> Publish Account.

Guests’ accounts can be added through your Azure Portal
Quickstart: Add guest users in the Azure portal – Azure AD | Microsoft Docs

2. Give the Publish account Read permissions on your Azure Analysis Service.
   You can do this by adding a new role, or just adding it as a server admin in the Azure portal.
   https://docs.microsoft.com/en-us/azure/analysis-services/analysis-services-manage-users

Now your LiveConnection is ready to go.

Webdashboard in a Pro Environment

In a Pro Environment the connection is made through a Service Principal. This can’t be done in a Standard Environment, because a Service Principal can only live in 1 Azure Active Directory Tenant. It’s the most secure way to connect to Power BI Embedded, because no User Account is needed.
For more information please read this: https://docs.microsoft.com/en-us/power-bi/developer/embedded/embed-sample-for-customers?tabs=net-core

Adding Webdashboard to your Pro environment is done with a Service Principal. All you have to do is give this Service Principal Read permissions to your Azure Analysis Server and you are ready to go.

Row Level Security (RLS)

Normally (for Import and DirectQuery) Webdashboard sends the users email property that is logged in to Webdashboard in the connection to the dataset. In your RLS implementation you can filter your data with the expression USERNAME().
LiveConnection defines it’s RLS roles in Analysis Services and not in Power BI. Next to that, the security is not only checked in Power BI, but also in Analysis Services. Which means that the Username must be an account that has direct access to the Analysis Model.
Solution: Webdashboard sends the Publish Account as the username and the email property of the user through CustomData. Now you can access the email of the logged in user with the DAX expression CUSTOMDATA().
Using row-level security with embedded content in Power BI embedded analytics – Power BI | Microsoft Docs

For more information about configuring this in your environment, refer to this manual:
Analysis Services LiveConnection manual