> I thought the list would be interested in this comparison of bonnie results:
>
> -------Sequential Output-------- ---Sequential Input-- --Random--
> -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
> Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
> 2.0.34 500 1117 16.2 1111 5.8 607 31.2 1426 92.7 1605 95.6 17.9 0.5
> 2.1.105 500 2208 21.8 2230 6.9 1214 24.3 3425 83.9 4280 82.1 26.8 0.7
I think there is more to it. I get decreasing (avg ~1-20%) bonnie
performance with 2.1.105 compaired to 2.0.3x but still 2.1.105 is
lightyears :) faster than 2.0 just in about anything - even the bare eye
can see it.
So basically what I wanted to say is: it is easy to do benchmarks but it
is quite tricky to interpret their results correctly.
Of course the results above would make perfect sense if there had been a
Adaptec driver update that would count for a 100% increase in
performance; which hasn't been the case for my ncr53c875.
-- Samuli Kaski, samkaski@cs.helsinki.fi Department of Computer Science, University of Helsinki, Finland.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu