[Linux-cluster] GFS RG size (and tuning)

[Jos Vos]

On Mon, Oct 29, 2007 at 08:53:10PM -0400, Glen Dosey wrote:

> We have several 3.3TB and larger filesystems on GFS1. I found that using
> the larger RG helps with performance. Increasing the statfs_slots seemed
> to help with df time and similar tasks as well. We have a filesystem
> with smaller RG's and I don't think the difference is worth moving the
> data around. If there is some other reason to make changes then I will
> increase the RG size. The statfs_slots can be increased on the fly and
> you should hopefully see an improvement. 

Unfortunately, no.  We see dramatic performace, df's may take 10+ minutes,
and above approx. 100 Mbit/sec total network traffic (mostly read from
GFS disks) the performance goes up and down, sometiems back to 5 Mbit/sec etc.
I have statfs_slots now set to 128, or should that be bigger?

We created now one fs with larger RG's and at least df works better, but
I fear we also have to do the other patches.

-- 
--    Jos Vos <jos at xos.nl>
--    X/OS Experts in Open Systems BV   |   Phone: +31 20 6938364
--    Amsterdam, The Netherlands        |     Fax: +31 20 6948204