Register now to learn Fabric in free live sessions led by the best Microsoft experts. From Apr 16 to May 9, in English and Spanish.
Hi all,
I have a P1 capacity and I need to perform some stress test in order to test the peak rated load and the predicted supported load. I would like to know how many users can open and refresh the same reports at the same time and which is the dependency of the cpu performance with respect to the number of connected users.
Can someone suggest me how to obtain such information (e.g. a tool in order to get these informations)?
Thanks in advance!
Solved! Go to Solution.
Hi @SampaSim,
Sorry @lbendlin, I don't totally agree. Because for measuring CPU for example, the load comes mainly from the users opening reports, mostly from (complex) DAX measures. Memory is, like you mentioned, consumed by datasets and refreshes running concurrently.
There is a great stress test tool available for Premium capacities, it's explained in detail in this blog post:
Power BI Premium – Know What Your Premium Capacity Can Handle.
The link to the GitHub repository is: https://github.com/microsoft/PowerBI-Tools-For-Capacities.
I'm assuming you already know/read this whitepaper? Otherwise you really should!
You can use the Power BI Premium Capacity Metrics app to filter by capacity, and review performance metrics for workspace content.
Paul Zheng _ Community Support Team
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Thank you all!
With the tool that you suggested, should I simulate something like an incremental load?
@SampaSim No, incremental refresh is designed to reduce load. You want more stress, not less.
@nickyvv Sorry, I should have qualified my remark - in our company we actively monitor the report performance and help developers reduce both visual rendering time as well as the number of visuals. Out of sheer necessity as most reports are used by hundreds of users. So in our setup user consumption of datasets that are already in memory plays a much smaller role than all the other actions.
You're measuring the wrong thing. The service doesn't care much how many users are looking at one dataset. What it cares about is concurrent dataset refreshes, dataset sizes and evictions. Bombard it with refresh requests for different large-ish datasets (1GB+ for your SKU) that you know will exceed the 25GB capacity when taken together. And then have different users request rendering of these different datasets at random.
Hi @SampaSim,
Sorry @lbendlin, I don't totally agree. Because for measuring CPU for example, the load comes mainly from the users opening reports, mostly from (complex) DAX measures. Memory is, like you mentioned, consumed by datasets and refreshes running concurrently.
There is a great stress test tool available for Premium capacities, it's explained in detail in this blog post:
Power BI Premium – Know What Your Premium Capacity Can Handle.
The link to the GitHub repository is: https://github.com/microsoft/PowerBI-Tools-For-Capacities.
I'm assuming you already know/read this whitepaper? Otherwise you really should!
Covering the world! 9:00-10:30 AM Sydney, 4:00-5:30 PM CET (Paris/Berlin), 7:00-8:30 PM Mexico City
Check out the April 2024 Power BI update to learn about new features.