Tyan Tiger 2460

Robert G. Brown rgb at phy.duke.edu
Mon Apr 29 08:44:19 PDT 2002


On Mon, 29 Apr 2002, Daniel R. Philpott wrote:

> Just a quick observation, looking at the different configurations you
> tried makes me think that the working configuration may have modified
> (slightly) the latency of the PCI bus.  Have you tried modifying the PCI
> latency in BIOS to work without the extra hardware?
> 
> I don't use the Tiger 2460 boards (I was a sucker for the S2462) so I
> can't test this hypothesis but maybe someone else can test it out.  

We did play with some of the slot settings in the bios to no avail.
We're in mid-test of the slot 2 riser solution (which has worked fine
over the whole weekend at a load average between 3 and 6 and with the
disk and network being heavily banged on all the while, so it looks like
it works).  As soon as it finishes (or I feel like grabbing another of
the idle chassis to test with and open it up) I'll give this a more
focused try.

It does have the feel of a latency issue, but why slot 1.  Why slot 1
when the system is (otherwise) entirely empty?  I'm not asking you,
understand, just looking to heaven for Enlightenment...

   rgb

> 
> Dan
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
> 

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu






More information about the Beowulf mailing list