Re: [Lse-tech] Re: [patch] scheduler fix for 1cpu/node case

From: Erich Focht (efocht@hpce.nec.com)
Date: Fri Aug 01 2003 - 11:30:07 EST


On Friday 01 August 2003 02:26, Martin J. Bligh wrote:
> np - was easy to fix up ;-) I did run some benchmarks on it ...
> low end SDET seemed highly variable, but otherwise looked OK.
> If I have only 4 tasks running on a 16x (4x4), what's the rate
> limit on the idle cpus trying to steal now?

the rate is 3ms, that's a bit too fast, maybe. And the busy rate is
200ms (instead of 400). I made some experiments with 2 CPUs per node
and 1 ms seems to be definitely too fast. I'll tune the formula a
bit...

Regards,
Erich

-
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 : Thu Aug 07 2003 - 22:00:16 EST