Comment 20 for bug 1833322

Hi Dough

> If irqbalance is to be included by default, then there should be due
> diligence to demonstrate a clear benefit.

You are right that we should have that as well.
But this would be even more ture if this would be about "making it the default
when it was not before".
Right now (purely opinion) the lack of data can IMHO neither be used to keep
it nor to remove it - which sadly locks this up a bit.

> The results were:

I want to thank you a lot, this won't be enough but it is a masterpiece
demonstration of dedicating time to start providing such data.
Thank you.

I do not know the ping pong test, but on iperf, I think that is in the noise
range as far as I remember. If you'd just re-run that as-is what is the delta
on your test box?

Hoping that this will be extended by more contributing different workloads
on different systems let me ask, what kind of system (cpu, size, nodes, ...)
was that. I know you are good at writing up things, you might set the standard
how others might report to this :-)

Your results show no change or minimal degradation while at the same time losing
a bit of power. Have you also had a chance to try the powerthresh argument
that Steve mentioned above?