We are using Azure Stream Analytics to build out a new IoT product. The data is successfully streaming to Power BI but there is no way to implement Row Level Security so we can display this data back to a customer, limited to only that customer's data. I am considering adding an Azure SQL DB between ASA and PBI and switching the PBI Dataset from a streaming dataset to Direct Query with a high page refresh rate but this seems like it will be a very intense workload for an Azure SQL DB to handle. There is the potential, as the product grows, for multiple inserts per second and querying every couple of seconds. Streaming seems like the better answer besides the missing RLS. Any tips?
There is the potential, as the product grows, for multiple inserts per second and querying every couple of seconds.
A small Azure SQL Database should handle that load. 1000/sec simple. 100,000/sec is probably too much.
And ASA can ensure that the output streams are not too frequent.