[Beowulf] CCL:Question regarding Mac G5 performance (fwd from mmccallum at pacific.edu)

Bill Broadley bill at cse.ucdavis.edu
Wed May 19 15:13:32 PDT 2004

> I had done some comparisons between price/performance, and I found dual  
> G5s to be at or near the best in price/performance, especially if  
> things are recompiled with the IBM compilers (between 8-10 % speed  
> increase over pre-compiled (with apple gcc) versions of NAMD2, and  
> using standard gcc for charmm).  I would expect that things like  
> gaussian03 run very well (I believe gaussian uses the IBM compilers for  
> macOS).  For MD, the speedup seems to be due to the on-chip square root  
> eval.

Any actual numbers would be very useful, ideally with the compilers,
compiler options, motherboard, and similar.  Did you compare 1 job per
node vs 1 job per node?  Or 2 jobs per node vs 2 jobs per node?  4 or 8
dimms?  Which kind of memory PC2700? PC3200?

> The built-in Gigabit enet is attractive, also, as charmm and NAMD scale  
> very well with gigabit, and it makes myrinet less price-effective (when  
> used on any platform, wintel included, see
> http://biobos.nih.gov/apps/charmm/charmmdoc/Bench/c30b1.html for  
> example).  I decided that dual G5 xserve cluster nodes with gigabit  

I come to a different conclusion based on those graphs.  In the first
graph myrinet improves by a factor of 2.6 (250 -> 95 seconds) from 2
processors to 8, where gige improves by only 20% (255 -> 210).  In the
second graph gigE gets SLOWER from 2 to 8 processors.  Do you think in
either case the 8 node (let alone 16) gige cluster would have better
price/performance then a myrinet cluster?

Seems like for applications like shown on that page you shouldn't
really bother with a cluster over 2 nodes with gigE, not many people
would be willing to pay a factor of 4 more for 20% or even negative

> switches were much more cost-effective for me than any other processor,  

Cost effective = price/performance?  Can you make any numbers available?

> especially any high-bandwidth specialty comm method (apple's gigabit  
> has a pretty low latency also).

Oh?  Can you share the apple gigabit numbers?  What is "pretty low"?

> Additional considerations for us were the BSD environment which is more  
> secure than windows, and the OS is arguably more stable and supported  

I'd agree with more stable and supported for use as a desktop, I'd
disagree with stable and supported as computational node.  OSX is the
new player on the block in this space.  Do you really think you would
get a good response from calling apple's tech support line when the
scheduler or network stack isn't performing to your performance

Certainly a very reasonable thing.

> It is my impression that opterons, PIVs, G5s all have their advantages,  

Agreed, thus the value in sharing performance actual results for
specific codes in specific environments.

Bill Broadley
Computational Science and Engineering
UC Davis

More information about the Beowulf mailing list