One of the most common scenarios for those who use Microsoft technologies and intend to publish its data into Power BI typically use SQL Server as a data source.
Power BI provides several ways to access SQL Server information, among them there are some that need simple configurations and others that require developments. Below we summarize the different possibilities that you have available:
Power BI has a wide range of connector to access to several types of data sources, these include a connector for Azure SQL Database and one for Azure SQL Data Warehouse.
To access to a SQL Server, whether it be an instance of Azure IaaS or On-Premises instance you can use Power BI Gateways. There is a Power BI Enterprise Gateway for enterprise deployments and a Power BI Personal Gateway for personal use. You can download it directly from Power BI portal and install in a server near your SQL Server.
For additional information about Power BI Gateways refer to https://powerbi.microsoft.com/en-us/gateway/.
Both these features support direct query option, that is SQL Server is queried when the user accesses the report or dashboard. Typically you prefer to schedule the frequency of the update. You can schedule updates every few minutes.
If you want to get a dashboard that receives continuous updates getting the effect of a live dashboard you need to perform simple developments.
To update a Power BI dataset you can use different Microsoft Azure functionality:
Windows Azure Event Hubs + Windows Azure Stream Analytics
Power BI REST API
Both techniques need configuration on Microsoft Azure side and a Windows service that resides in the same network of your SQL Server that is responsible for reading the information and send them to the Cloud.
This approach allows you to make updates with a frequency close to the second.
Use Windows Azure Event Hubs and Microsoft Azure Stream Analytics is simpler and the Power BI dataset is created automatically, but with this technique, you can only insert new information.
The Power BI APIs are a bit more complex but allow several types of operations. For example it is possible to empty the dataset before inserting new information.
However, you must manually create the data set.
You can download a Windows service sample at the address http://db2powerbi.codeplex.com/. This simple solution it able to use both techniques. It is enough to apply a configuration to the QueryConfig.json file and write a query file. The project includes a console (DbToP Svc.Power Bi.UI.exe) that allows you to create the dataset and retrieve its id (required in the configuration file).
Below is an example of Query Config.json file.
As I said, before starting the service you need to apply some Microsoft Azure side configurations.
You can follow the article https://azure.microsoft.com/en-us/documentation/articles/stream-analytics-power-bi-dashboard/ to configure Windows Azure Event Hubs and a Stream Analytics Job.
You have to follow the article https://powerbi.microsoft.com/en-us/documentation/powerbi-developer-register-a-client-app/ before use Power BI REST API.