INTEL XEON vs AMD ATHLON

[Steven Timm]

On Wed, 12 Jun 2002, Ivan Oleynik wrote:

> I sent email a couple days ago asking for advice re linux cluster vendors
> and critical configuration issues (I am going to summarize my findings
> after getting all the responses). It turned out that the choice of the
> processor is the most critical one. I had an impression from literature
> that Athlon MPX goes head by head with Xeon in terms of CPU speed for
> heavy numerics, but it is considerably less expensive.
>
> Surprisingly, I received several responses that contradict to my view. In
> particular, Xeon is only 5-10% more expensive,

We just got a set of bids back for a large cluster... the Xeon
alternatives came in 20% higher than the Athlon alternatives... it
is not only the cost of the chips you have to consider, the boards
are more expensive as well.

> but it is more powerful
> since it is currently running at 2.4 GHz, has 400 MHz front bus plus 512
> Kb of secondary cash as compared to 1.76 GHz, 266 MHz and 384 Kb for
> Athlon. Also, the newest linux kernel will provide multithreading on Xeons
> that could also result in substantial boost for numerical calculations.
>

Key word:  could.. "It depends on your application"
> I think this is old question, but I would like to know what experts
> say: Is it really true that the new Xeon is superior than Athlon as far as
> heavy numerical calculations in cluster environment are concerned?
>

We find that AMD MP1800+ are superior in performance to Xeon 2.0 GHz
for the code mix we run, which admittedly has a lot more integer
content than the average high-performance application.
If you look at the SpecINT2000 and SpecFP2000 numbers the Xeon
is superior by quite a large margin...but to really get that
performance you have to (1) run a proprietary heap manager
(2) run Windows, since all the benchmarks are done in Windows,
(3) use Intel's proprietary compiler and the specific instruction sets
for the chips.

This (3) is not an option for us given that we are a world-wide
collaboration that tries to use open-source compilers wherever
possible.  Until the open-source compilers such as gcc catch up
to the new instruction set, the Intel Xeons are not worth our money.

Steve Timm


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