Determining Capacity for very large Report runs

[W_G_Ryan]

I have a scenario where each month, we need to generate roughly 1 million reports (account statements),  We're using  Power BI Report Builder and they'll be exported to PDF where we will store them in Azure Blob Storage.  We've got the process working now for one subset of accounts, which amounts to about 3k a month, but they're executed on demand and not in batch.  

We have  a P3/A6 capacity but are trying to figure out if this is going to be enough. Obviously things start getting very expensive if you move up but we also need this to run and process as quickly as possible (right now, it's done on Prem which is notably faster and takes about 7-10 days).  

Has anyone run into scenarios like this, high volume report rendering (we only need the PDF if that matters)?  We're going to start load testing but our test environment doesn't have the same capacity as prod so that's going to take some orchestration, and also slighly worried that firing a huge volume of requests might look like a cyber attack of some sort and I'm just looking to get ahead and anything I can.

Any and all advice is greatly appreciated.

[GilbertQ]

Happy to help!

[GilbertQ · Accepted solution]

Hi @W_G_Ryan 

What you could do is to rather spin up an Fabric Capacity when you want to run the reports on a monthly basis. When using the Fabric Capacity you can scale it up and scale it down as needed, as well as pause when not being used. So essentially. what that means is you could spin up a very large fabric capacity at the side of the month. let itrun through all your reports and then pause it or make it to a smaller capacity for the rest of the month.

[W_G_Ryan]

Hey there Gilbert:

That actually sounds perfect.  I thought you had to buy the capacity long term but that's awesome to know and definitely going to give it a try.