MIGUEL EGEA on 28 Mar 2024 15:31:45
When we create monitoring solutions, sometimes we need to be able to extend the functionality that the Capacity Metrics app already has. Events such as throttling, smoothing, and other penalties are not exposed via the API and are not exported as events to Log Analytics
What I propose as an idea is that these events be exposed in some way, ideally through Log Analytics as an additional event. The business motivation is as follows:
In a client scenario where capacity is shared among departments, it’s possible that one department has uploaded reports that result in excessive capacity consumption. However, it’s other departments that bear the penalty in terms of cost. To determine whether scaling is necessary, to request additional capacity, or even to separate certain processes into capacities that can be turned on and off, we need to have that information in the same way that the app’s semantic model already obtains it.
Ideally, it would be great if the exported event includes references to the queries or time intervals that caused that penalty, so that we can distinguish who is causing the problem and who is experiencing it