[ale] Dual core CPUs in cluster?

[James P. Kinney III]

On Tue, 2005-04-26 at 12:59 -0400, Dow Hurst wrote:
> Anyone have advice on whether there are issues with parallel 
> applications in a dual core vs single core cluster environment?  I have 
> an option of getting dual core 275 Opterons vs single core 248 
> Opterons.  Will be using the Ammasso 1100 interconnect for low latency.  
> 20 nodes if single core or 14 nodes if dual core based on pricing.  Just 
> wondering if anyone has any thoughts?  Main application is NAMD and 
> secondary is Schrodinger's Jaguar.  I have access to the NAMD cvs and 
> will have Pathscale compilers on the cluster.  Thanks,

Call me when you are ready to play :)

I suspect that dual cores will be good for situations that require very
closely coupled looping caluculations. The latency between cores should
be _really_ low. 

However, the datasheets I have looked at show something I don't like,
shared L1 and L2 cache. I don't recall if this is entirely accurate or
not or if it was a "special" version. It seems to me that a shared L1 is
asking for trouble. Shared L2 (the deepest data cache) would also be a
problem as both core would not be working on the same data thus
requiring a page out.

Hmmm.

> Dow
> 
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://www.ale.org/mailman/listinfo/ale
-- 
James P. Kinney III          \Changing the mobile computing world/
CEO & Director of Engineering \          one Linux user         /
Local Net Solutions,LLC        \           at a time.          /
770-493-8244                    \.___________________________./
http://www.localnetsolutions.com

GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics)
<jkinney at localnetsolutions.com>
Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part