The testing/staging environment should perform regular high load tests and monitor memory and CPU usage

The idea is to detect how the application will behave in situations where memory and CPU limiters are imposed by the infrastructure and properly address such issues. Also to fine tune the values we use for buffer sizes, and so on...

To be considered during test buffered channel sizes, number of workers, SQL operations performance, log processing performance, memory and cpu usage, and so on.

Ideally it should output some reports that we can use as evidence for finding and fixing regressions.