How can we forecast expected system utilization?

[notcool11]

Are there documents that help describe the expected sizing/scaling of the pods based on volume/utilization? There are tangible cloud costs for running this app, we need some way to know how to plan for those costs.

Also, without knowing what the expectations are, it is difficult to know what is normal and what needs to be addressed. Looking through the docs, there are recommendations to set up autoscaling but there is nothing that tells us how much scaling will happen based on usage.

Thanks!