Cause Messaging built into the Premium Capacity Metrics App/Report
Although the new Prem Cap Metric app is a good start to allowing organizations to understand and manage their Premium Capacity, it has a way to come.
Trying to get to the root cause of issues in capacity utilization is still a difficult puzzle. Better ways to help point the administrators to the issues is needed. As we know, there are several reasons for high utilization, some can be addressed in development (query tuning) and some operationally (staggering refreshes). And when those items are handled, the next step is capacity expansion.
I would like to see more details about the datasets in order to achieve root cause
1. metrics on interactive queries per dataset
2. metrics on refreshes that "time out" and what caused the time out
3. better controls to manage time out values so that we can better control how our resources are used, i.e. max values on connection times, refresh times, etc. This could keep one bad dataset from spoining the whole environment
Thanks for your suggestion. We are working to make improvements in this area.