decent performance from G4 Macs?

Iwao Makino rickey-co at mug.biglobe.ne.jp
Sat Apr 13 11:57:27 PDT 2002


Mark,

At 14:18 -0400 13.04.2002, Mark Hahn wrote:
>I'm doing some benchmarks to evaluate whether current Macs
>would make suitable nodes for a serial farm (lots of nodes,
>preferably fast CPU and dram, but no serious interconnect.)

I agree about first 2, but for Interconnect, there's Myrinet
for MacOS X!!!
Myricom now released MPICH-GM for MacOS X as well.

I personally haven't purchased Mac to test, but having said
by Apple that G4's with Blast is a lot faster than P4, I'm
quite interested to evaluate them soon.

>I've tried a variety of real codes and benchmarks, but can't
>seem to get something like a Mac G4/800 with PC133 to perform
>anywhere close to even a P4/1.7/i845/PC133.
>
>I'm using either the gcc 2.95 that comes with OSX or a
>recent 3.1 snapshot (which is MUCH better, but still bad).

I think you have to MODIFY code a bit to take advantage of velocity engine
for MacOS X gcc.

I thought there are interesting post along with other bluff.
<http://www.apple.com/pr/library/2002/feb/07blast.html>


>is it just that the performance Apple brags about is strictly
>in-cache, and/or when doing something ah specialized like
>single-precision SIMD (altivec/velocity engine)?

I too think that's big part of it...

-- 

Best regards,

Iwao Makino
Hard Data Ltd. Tokyo branch
mailto:iwao at harddata.com
http://www.harddata.com/

-> HPC cluster specialist<-
-> Scientific Imaging/Life Science/Physical Science/Parallel Computing <-



More information about the Beowulf mailing list