Re: [Linux-cluster] Slowness above 500 RRDs

[David Teigland <teigland redhat com>]

On Wed, Apr 25, 2007 at 05:47:53PM +0200, Ferenc Wagner wrote:
> David Teigland <teigland redhat com> writes:
> 
> > On Mon, Apr 23, 2007 at 04:17:18PM -0500, David Teigland wrote:
> >> > Also, what's that new infrastructure?  Do you mean GFS2?  I read it
> >> > was not production quality yet, so I didn't mean to try it.  But again
> >> > you may have got something else in your head...
> >> 
> >> GFS1 and GFS2 both run on the new openais-based cluster infrastructure.
> >> (in the cluster-2.00.00 release, and the RHEL5 and HEAD cvs branches).
> >
> > I've attached a little flock/plock performance test that emulates what
> > you're doing; could you run it on your cluster and send the results?
> 
> Here you go.  This is with three nodes mounting the FS, one running
> the test.  /proc/cluster/lock_dlm/drop_count was set to 0 before
> the mounts.

Thanks, some of these numbers look odd to me; I'll need run the test on my
own rhel4 cluster to understand them better.

In the end, though, I think you'll get the best performance by using
flocks under the new cluster infrastructure and kernel.

Dave