opteron VS Itanium 2

Richard Walsh rbw at ahpcrc.org
Thu Oct 30 10:57:00 EST 2003


On Wed Oct 29 21:38:48 2003, Mark Hahn wrote:

>> >To sum up, the question is: is the Itanium2 worth the price difference
>> >or is the Opteron the best choice?
>> 
>>  The SpecFP2000 performance difference between the best I2 and best
>>  Opteron seems to be about 600 spec points or 40% (~1400 versus ~2000).
>
>which to me indicates that the working set of SPEC codes is a good 
>match to the cache of high-end It2's.  this says nothing about It2's,
>but rather points out that SPEC components are nearly obsolete
>(required to run well in just 64MB core, if I recall correctly!)

Of course, there is some truth to what you say, but "this says nothing about 
It2" seems a tad dramatic (but ... definitely in character ... ;-) ).  Below is 
the memory table for most of the benchmarks.  A few fit in the 6MB cache (although
some surely should, as some codes do or can be made too fit into cache).  Many 
are in the 100 to 200 MB range.  The floating point accumen of the I2 chip is hard 
to question with the capability of performing 4 64-bit flops per clock (that's 
a 6.0 GFlops peak at 1.5 GHz; 12.0 at 32-bits).  Moreover, even an I2 with 1/2 the 
Opteron's clock and only 50% more cache (L3 vs L2) performs more or less equal to the 
Opteron 246 on SpecFP2000.  

And after all a huge cache does raise the average memory bandwidth felt by the 
average code ... ;-) (even as average codes sizes grow) ... and a large node count 
divides the total memory required per node.  Large clusters should love large caches
... you know the quest for super-linear speed ups.

The I2's weakness is in price-performance and in memory bandwidth in SMP configurations
in my view. My last line in the prior note was a reminder to the original poster
that SpecFP numbers are not a final answer.  I repeated the "benchmark you code"
mantra ... partly to relieve Bob Brown of his responsibility to do so ;-). 

Got any snow up in the Great White North yet?

Regards,

rbw


            max     max     num  num
            rsz     vsz     obs  unchanged  stable?
            -----   -----   ---  ---------  -------
gzip        180.0   199.0   181      68
vpr          50.0    53.6   151       6
gcc         154.0   156.0   134       0
mcf         190.0   190.0   232     230     stable
crafty        2.0     2.6   107     106     stable
parser       37.0    66.8   263     254     stable
eon           0.6     1.5   130       0
perlbmk     146.0   158.0   186       0
gap         192.0   194.0   149     148     stable
vortex       72.0    79.4   162       0
bzip2       185.0   199.0   153       6
twolf         3.4     4.0   273       0

wupwise     176.0   177.0   185     181     stable
swim        191.0   192.0   322     320     stable
mgrid        56.0    56.7   281     279     stable
applu       181.0   191.0   371     369     stable
mesa          9.4    23.1   132     131     stable
galgel       63.0   155.0   287      59
art           3.7     4.3   157      37
equake       49.0    49.4   218     216     stable
facerec      16.0    18.5   182     173     stable
ammp         26.0    28.4   277     269     stable
lucas       142.0   143.0   181     179     stable
fma3d       103.0   105.0   268     249     stable
sixtrack     26.0    59.8   148     141     stable
apsi        191.0   192.0   271     270     stable


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



More information about the Beowulf mailing list