ATHLON vs XEON: number crunching

Ivan Oleynik oleynik at chuma.cas.usf.edu
Fri Jun 21 09:16:06 PDT 2002


Thomas,

As I said in my original email I used PGI compilers (the code is a mixture
of F90 and C), it uses MPI, not OpenMP.

This morning I already mentioned about my new tests, when I loaded both
processors in both Athlon and Xeon nodes. The load was done by running
mpirun -np 2. Both Myrinet and Gigabit ethernet gave approximately the
same timing which is what you would expect in the absence of internode
communication.

I could not make comparison of scalability because I don't have more than
1 node available.

My results remain unchanged: Xeon 2.2 GHz is 1.5 times faster than Xeon XP
2100+.

Ivan

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

On Fri, 21 Jun 2002, Thomas Guignon wrote:

> 
> Hello
> I would be interesting to know which compiler was used for this stream 
> benchmark? probably is the Intel compiler.
> It's possible to improve the AMD numbers if prefetch is not in use but it 
> will change the scalability conclusion of  Ivan Oleynik
> The fact that dual Xeon perf decrease is quite strange (no speedup would be 
> more normal) and there is some investigation to do:
> - Does the  2 processors test was obtained with openMP (I guess it is) ?
> 

------------------------------------------------------------------------
Ivan I. Oleynik                       E-mail : oleynik at chuma.cas.usf.edu
Department of Physics
University of South Florida
4202 East Fowler Avenue                  Tel : (813) 974-8186
Tampa, Florida 33620-5700                Fax : (813) 974-5813
------------------------------------------------------------------------




More information about the Beowulf mailing list