| 1 |
Unfortunately BPS has an annoying habit of overwriting the test |
| 2 |
output files every time it is run regardless of which test you |
| 3 |
actually run. I didn't notice this when I was doing my initial stuff |
| 4 |
on RedHat so I lost all the other tests results from BPS. I have |
| 5 |
better tests of the hardware since then. We ran HPL on the PowerPC |
| 6 |
cluster running both Gentoo 2005.1 and Mac OS 10.4 (mac os was ~ %5 |
| 7 |
faster on average). I ran IOZone on our XRaid using HFS+ under Mac |
| 8 |
OS, and Reiser, xfs, jfs, and ext3 under Gentoo for PPC64. |
| 9 |
|
| 10 |
I also ran tests using Mr Bayes, ClustalW, MPIBench, and several |
| 11 |
other in house tests that I had available but all of these tests |
| 12 |
where either inconclusive or failed to demonstrate any difference |
| 13 |
between the platforms. |
| 14 |
|
| 15 |
I will be doing a bunch of this again here in a bit in order to get |
| 16 |
data on our new OS build we are putting into production =) |
| 17 |
|
| 18 |
|
| 19 |
On Apr 11, 2006, at 6:20 PM, Brandon Edens wrote: |
| 20 |
|
| 21 |
>> Yeah, from the numbers it looks as if it would be dependent on the |
| 22 |
>> purpose of the cluster whether OS X or Gentoo would do better. On |
| 23 |
>> ppc, |
| 24 |
>> Gentoo does poorly on the first two benchmarks and also on context |
| 25 |
>> switching. On x86, the first two are more comparable with RH, but the |
| 26 |
>> others, Gentoo has a small to large advantage over RH, just as on |
| 27 |
>> ppc. |
| 28 |
> |
| 29 |
> First and second benchmarks are dhrystone and whetstone, synthetic |
| 30 |
> benchmarks. |
| 31 |
> I'd be wary of these types of benchmarks; they've been cheated |
| 32 |
> before. I'd like |
| 33 |
> to see some real-world computation benchmarks. |
| 34 |
> |
| 35 |
> Brandon Edens |
| 36 |
> |
| 37 |
|
| 38 |
-- |
| 39 |
gentoo-cluster@g.o mailing list |
Archives