I know that there are various ways of inspecting the evaluation loop e.g.:
Average of last 10 cycles - average duration of evaluationcycle of client
Get the “heavy hitters” - track fixlets of evaluationcycle of client
Is there a way to get a “breakout” of the average loop times? i.e. Can I see the actual 10 results that are use to build the average?
I’m looking at a machine with a current average of 2-3 hours, but the usage profile data doesn’t seem to support it so breaking out the averages would be very useful.
And yes I’m getting obsessed by loop times at the moment. I’ve managed half the average times on our current estate, but it’s still way too high!