Lee Revell wrote:
On Fri, 2005-08-19 at 05:09 +0200, Michal Piotrowski wrote:
Hi,
here are interbench v0.29 resoults:
The X test under simulated "Compile" load looks most interesting.
Most of the schedulers do quite poorly on this test - only Zaphod with
default max_ia_bonus and max_tpt_bonus manages to deliver under 100ms
max latency. As expected with interactivity bonus disabled it performs
horribly.
I'd like to see some results with X reniced to -10. Despite what the
2.6 release notes say, this still seems to make a difference.
Even spa_no_frills, which does absolutely nothing to help interactive
(or other special interest) tasks, can cope in these circumstances as
illustrated by these results from my (relatively old) SMP machine show:
--- Benchmarking simulated cpu of X nice -10 in the presence of
simulated ---
Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met
None 0.01 +/- 0.129 2 100 99.3
Video 0.007 +/- 0.0818 1 100 99.3
Burn 0.006 +/- 0.0817 1 100 99.3
Write 0.033 +/- 0.271 3 99.3 98
Read 0.046 +/- 0.337 3 98.4 97
Compile 0.023 +/- 0.208 2 99.3 98.3
Memload 0.043 +/- 0.31 3 98.1 97
This machine isn't directly comparable with Michal's so for comparison
here are the results from "out of the box" Zaphod on the same machine:
--- Benchmarking simulated cpu of X in the presence of simulated ---
Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met
None 0.02 +/- 0.2 2 99.3 98.7
Video 0.007 +/- 0.0818 1 100 99.3
Burn 0.023 +/- 0.208 2 99.3 98.3
Write 0.147 +/- 0.949 12 94.7 93.2
Read 0.033 +/- 0.258 2 98.7 97.7
Compile 2.94 +/- 10.7 105 76.8 71.6
Memload 0.017 +/- 0.153 2 100 98.7
As you can see there's evidence in these numbers the file writes are
implicated in the bad numbers for the Compile load (which is a mixture
of Burn, Read, Write and (I think) Memload). So testing with different
I/O schedulers might be interesting.
Peter
--
Peter Williams [EMAIL PROTECTED]
"Learning, n. The kind of ignorance distinguishing the studious."
-- Ambrose Bierce
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
