I've updated the contest responsiveness benchmark thanks to Rik Van Riel. It now
works on 2.5.x kernels. Here is an updated set of benchmarks:
No Load:
Kernel Time CPU
2.4.19 1:07.69 99%
2.4.19-ck7 1:07.62 99%
2.5.34 1:09.97 99%
2.4.19-ck7-rmap 1:08.83 99%
2.4.20-pre7 1:08.53 99%
CPU Load:
Kernel Time CPU
2.4.19 1:20.59 80%
2.4.19-ck7 1:10.79 93%
2.5.34 1:12.09 94%
2.4.19-ck7-rmap 1:11.07 93%
2.4.20-pre7 1:22.37 80%
IO Load:
Kernel Time CPU
2.4.19 1:50.85 58%
2.4.19-ck7 1:17.07 85%
2.5.34 1:18.43 86%
2.4.19-ck7-rmap 1:21.48 81%
2.4.20-pre7 1:54.14 58%
Mem Load:
Kernel Time CPU
2.4.19 1:24.94 77%
2.4.19-ck7 1:11.36 93%
2.5.34 1:20.65 86%
2.4.19-ck7-rmap 1:40.37 68%
2.4.20-pre7 1:28.10 76%
The absolute values are not important, only their differences. The smaller the
difference (in time) between noload and the different loads the better. The
higher the cpu percentage, the better.
I'm working on updating the IO load for the next version to something a bit more
stressful on the system thanks to Andrew Morton's suggestions.
The current version is 0.12 and can be downloaded at my kernel page:
http://kernel.kolivas.net
Cheers,
Con.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Sep 15 2002 - 22:00:37 EST