[Bug 173541] High (0.60+, 1.00) idle load averages

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=173541

--- Comment #39 from Stavros Filargyropoulos <stav@stav.cc> ---
(In reply to Mark Millard from comment #38)

> I still do not see how this is to be classified as unreasonable
for the background activity present, even if it was an arithmetic
mean.

One argument is too look what 9.0-RELEASE used to report. Which would be an
order of magnitude less.

Another argument is to try to visualize what a load of 0.20 would mean if we
had a single process and a very simple kernel that wouldn't run anything other
than idle/scheduling and that process.

Ignoring the more complex computation and taking just the arithmetic mean. 0.20
load would mean that we would have a loop running, since it's the only process,
for 20% of a minute.  That means it would run for 12 seconds and sleep for 48.

I don't think that background activity is worth 12 seconds per minute. FreeBSD
would have serious performance issues if it did.

Yet another argument is to run something equivalent to the background activity.
If things were fine and it's just our perception that is wrong, then I would
expect the load to double. It doesn't..

-- 
You are receiving this mail because:
You are the assignee for the bug.