SQL Server provisioned or serverless for stream analytics data

Occasional Visitor

This is my first message to this great community.

 

I want your guidance on the following scenario:

  • our company has clients that have devices (sensors mostly) that stream data to Azure's IoT hub.
  • These data are stored to a SQL database in Azure and retrieved (through Direct Query) from Power BI reports and dashboards.
  • The above process takes place almost in real time.

For the moment SQL server is configured as serverless, but we think that the cost is high.

 

I would love some help according the choice between serverless and provisioned. Which is the best one according the above IoT scenario? And how we can calculate that cost in relation with stream analytics queries?

 

Thanks in advance

0 Replies