[Gllug] Historical performance reporting

[John Hearns]

2009/3/28 Stephen Nelson-Smith <sanelson at gmail.com>:
>
> Sorry - all I meant was I was finding it hard to translate into a
> non-clustered environment, but I think I've got it now - I just think
> of groups of servers, eg webservers, or rhel4 servers as a cluster.

That's really all a Beowulf type cluster is.

>>> Suppose in my lab I set up 3 VMs, and one for running ganglia - how
>>> would I go about gathering info on those 3 vms?
>>
>> a) run the gemetad daemon on the 'one for running ganglia'
>>
>> b) run gmond on the 3 VMs
>
> Yep - got all that working - I must say I'm very impressed indeed.
> Anyone runnin gmond on Solaris and Windows machines, and aggregating
> on Linux?  That's the next thing to test.
That would be interesting - I've never seen that done.
-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug