Re: some more performance benchmarks

Neels Hofmeyr wrote on Thu, Apr 07, 2011 at 16:42:43 +0200:
> I have no idea about other loads on our vm. Maybe we could use the
> user/sys output of 'time -p' instead of the real-world time measurements
> it is taking right now, although my intention was to stick to the actual
> real world time. Of course that works best on an otherwise idle system.
> danielsh, or anyone, do you have any idea at what times-of-day (or
> something) the VM host is most likely to idle for ~2 hours, so we can
> schedule the tests then?

It's two-fold:

* When is the VM itself inactive
* When is the host inactive

The first question is easy: just tell artagnon and danielsh "don't use
svn-qavm at the moment". Just edit /etc/motd, or edit /topic on IRC,
or something.

The second question is harder. The host runs over 30 other VMs, and
I am not sure offhand what they are all used for. If you care about
this level of detail, you'd have to talk to Gavin or Tony from infra ---
try catching them on infrastructure@ or #asfinfra.
Received on 2011-04-08 01:05:20 CEST