NFS Performance (was Re: [Beowulf] GPFS on Linux (x86))

Michael Will mwill at penguincomputing.com
Fri Sep 15 11:36:24 EDT 2006


I am puzzled by an sles9sp3 (2.6.9 kernel) nfs server that serves rhel3 (2.4.21 kernel) compute nodes. For some reason a lot of times the mounts fail (with default as well as modified parameters). The symptom is mount: rpc timeout. The server logs all authentification requests as successful.  The switch is an oversubscribed hp 4108gl.

Google turned up nothing useful.

Any ideas?

Michael

 -----Original Message-----
From: 	Brent Franks [mailto:brentasterisk at gmail.com]
Sent:	Fri Sep 15 07:47:11 2006
To:	Chris Samuel
Cc:	beowulf at beowulf.org
Subject:	Re: NFS Performance (was Re: [Beowulf] GPFS on Linux (x86))

On 9/15/06, Chris Samuel <csamuel at vpac.org> wrote:
> Wild guess - NFS servers are running RHEL and using ext3 ?

Exactly.

> We used to have that problem with the 2 NFS servers for our 90 node cluster
> till we dumped RHEL,  switched to Fedora and XFS and all our horrible NFS
> problems went away.

Nice, any sort of comparison data in terms of differences in
throughput achieved?

Additionally, are you writing your journal to a different partition?

Thanks,

- Brent
_______________________________________________
Beowulf mailing list, Beowulf at beowulf.org
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.clustermonkey.net/pipermail/beowulf/attachments/20060915/2524dd0a/attachment-0001.html>
-------------- next part --------------
_______________________________________________
Beowulf mailing list, Beowulf at beowulf.org
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf


More information about the Beowulf mailing list