Re: [Condor-users] tuning a file server

[David McBride <dwm@xxxxxxxxxxxx>]

On Tue, 2005-08-02 at 12:44 +0200, Steffen Grunewald wrote:

> > The 2.6 IO schedulers tend to _better_ than the 2.4 one.  The default is 
> > to use the 'anticipatory' scheduler which tends to be excellent for most 
> > needs.
> 
> I suppose it's anticipating the wrong stuff. If 100+ client tasks start
> to send off 24 nfsd threads to work, probably the scheduler gets a bit
> confused. I guess that some elevator type would be the better choice?
> (Can you point me to some docs?)

http://www.cs.rice.edu/~ssiyer/r/antsched/
linux-2.6.12.2/Documentation/block/as-iosched.txt

Aha! The latter has some relavant comments that I wasn't aware of:

"Attention! Database servers, especially those using "TCQ" disks should
investigate performance with the 'deadline' IO scheduler. Any system
with high disk performance requirements should do so, in fact."

[...]

"Also, users with hardware RAID controllers, doing striping, may find
highly variable performance results with using the as-iosched. The
as-iosched anticipatory implementation is based on the notion that a
disk device has only one physical seeking head.  A striped RAID
controller actually has a head for each physical device in the logical
RAID device."

Cheers,
David
-- 
David McBride <dwm@xxxxxxxxxxxx>
Department of Computing, Imperial College, London

Attachment: signature.asc
Description: This is a digitally signed message part