[MITgcm-support] AMD quad core

Dierk Polzin dtpolzin at wisc.edu
Tue Jan 6 14:15:54 EST 2009


I see now if you go back far enough you can get a lot of responses  
about this topic...
hmmmmmmm

-----------------------------------

http://mitgcm.org/pipermail/mitgcm-support/2008-April/005331.html

http://forge.csail.mit.edu/pipermail/mitgcm-support/2007-October/ 
005076.html

http://mitgcm.org/pipermail/mitgcm-support/2008-October/005731.html

http://mitgcm.org/pipermail/mitgcm-support/2008-October/005738.html

http://mitgcm.org/pipermail/mitgcm-support/2008-October/005748.html

http://mitgcm.org/pipermail/mitgcm-support/2008-October/005749.html


=======================================================
On Mon 14 Apr 2008, Michael A. Spall wrote:

 > I have been told that the MITgcm does not scale well with
 > multicore processors.  I am thinking about say 16 processors on
 > 2 or 4 nodes (2 processors per node, dual or quad processors) with
 > infiniband.  Does anyone have experience with this?

So far in our testing at least dual core scales OKish - ie. on both  
Intel and
AMD systems we're scaling great with 2 cores out of 4 running user  
code but
see less than 3 speedup going to all 4 cores. I have not tried to test
running on 3 out of 4 cores on a node yet but I doubt it would be  
greatly
worth it because the problem is with insufficient memory bandwidth  
for memory
streaming at the socket level for AMD architectures and at the FSB  
per socket
level for the Intel architectures.

Given that, quad cores are unlikely to be worth it for Intel systems.  
AMD
Barcelona might be a somewhat better option according to published  
memory
performance numbers but I have not run on it yet - Chris has. However  
I don't
expect to see more than a speedup of 3 within each quad core  
processor or 6
in a dual socket node.

On the other hand you'll soon probably be out of options as all  
processors are
going to be quad core or more and dual core "HPC" systems may not  
catch on
because of lack of volume.

Constantinos
-- 
=======================================================




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20090106/f8a2f0ff/attachment.htm>


More information about the MITgcm-support mailing list