One of our enterprise-level customers reached the Power BI capacity limit on dedicated cloud computing and storage resources. They already owned the Power BI Premium licensing package and didn’t want to spend more money to buy additional resources.
What to do when reaching your Power BI Premium limit
Reaching the resource capacity limit in Power BI may lead to serious consequences. Data models using this service will fail when refreshing through the gateway. The business will not get new data, causing all teams to fall behind on analysis of information.
Typical solutions and their limitations
The first approach to the above-mentioned problem is to buy more capacity. Adding another Power BI Premium license will cost $4,995 per month. This can get pricey and is the reason most of our customers don’t want to invest in a second premium license. Of course, there is always the possibility these new resources will quickly max out again.
The second approach is to prevent Power BI models from using capacity resources. In this case, the removed models will not refresh on schedule. As a result, users will need to update their data manually. But in large organizations, everyone believes their report is a priority. So, how can you reasonably determine which schedules to turn off?
The third typical approach is to optimize the environment to use fewer resources. Once again, you need to wisely determine which reports need optimizing. And then you would need to train users on best practices for optimizing their reports.
Overall, these three approaches look quite complicated and unsustainable.
DB Best’s solution
But there is a better approach. First, we optimized current reports to use fewer resources.
My secret weapon of choice is the Power BI Premium Capacity Metrics tool. My approach is to analyze the top poor performing reports and look at ways to optimize them. Finally, I’ll set up a Power BI jam session with the report authors and show them how to make the optimization recommendations.
Go on…admit it, we all have test datasets lying around our workspaces. When things get tight, I look at removing test models from Premium Capacity and instead place them in shared spaces (which are not using resources). This allows for the creation of a typical development/production architecture inside the Power BI service. Using two workspaces instead of one allows users to refresh reports without using premium resources.
At the end of the day, I’ll work together with our customers to craft a plan for monthly capacity clean up. This approach allows a continuous practice for optimizing reports to use fewer resources.
This is how I and the other Power BI consultants at DB Best help our customers save money and resources after reaching the Power BI Premium limit.
