[Beowulf] Keeping the Athlon MP cluster limping along

Robert G. Brown rgb at phy.duke.edu
Thu Dec 9 12:12:06 PST 2004


On Thu, 9 Dec 2004, David Mathog wrote:

> Besides, it may well be that the CPUs are bad and the
> motherboards are all right.  Not having any spare known
> good CPUs or known good motherboards there's no way to
> play mix and match to figure out which component is the problem.
> Well, not unless we sacrifice one of the working nodes, and
> I'm really hesitant to do that in case it's one of those horrible
> situations where component A breaks component B, which would
> result in us having 3 flakey nodes instead of 2.  In any case, if
> the CPUs are bad they'll be around $150 to replace from a vendor
> and the motherboard about $190 (somewhat less on Ebay, but that's not
> how I want to buy components.)

Both AMD and Tyan have been pretty decent about fullfilling the terms of
their 3 year mfrs warranty on both the CPUs and motherboards.  So before
you throw anything away, most definitely look into RMAing broken parts
less than 3 years old.  If nothing else, you could put replaced
processors into the second socket of good motherboards, and keep
replaced motherboards around as spares.  If you kept one "known good"
system more or less powered off, it would give you a testbed of sorts,
although we see problems with heating fans and more that only surface
(as you note) under heavy or variable load in a sealed case and not open
running just a single exerciser on the bench.

   rgb

> 
> Regards,
> 
> 
> David Mathog
> mathog at caltech.edu
> Manager, Sequence Analysis Facility, Biology Division, Caltech
> _______________________________________________
> 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