[testers wanted] schedule bigpatch

Rik van Riel (H.H.vanRiel@phys.uu.nl)
Sun, 6 Dec 1998 16:17:06 +0100 (CET)


I am currently looking for volunteers who want to evaluate
the performance of the scheduler bigpatch on LARGE multi-
user machines.

The scheduler bigpatch contains a number of items -- all of
which have proven to work stable under the most obscene
loads -- that improve the Linux scheduler with respect to
interactive respons, realtime response and overhead.

The particular feature that might contain a few performance
thingies is the improvement of interactive response. The
(often large) improvement is achieved through load-independant
'recharging' of the priorities of processes that have slept
for a period.

I have a slight suspicion that on machines where most of the
load is generated by interactive programs (100+ people reading
mail at the same time) the load-independantness might actually
degrade performance a bit compared to what it could have been
with an optimal scheduler.

Since it's all about _perceived_ performance testers don't
need to do any complicated measurements at all, just tell
me your impressions...

The scheduler bigpatch is available for kernel versions
2.1.125 to 2.1.131. If you have always been afraid of
development kernels, this is the time to make the switch
to "pre-2.2" 2.1.131, which is as stable as 2.0 will ever

You don't have to be afraid about the stability of my
patch either -- it's been tested (read: severely abused
and stressed to infinity) on quite a lot of systems for
the last two months now.

You can get the patch from: http://www.phys.uu.nl/~riel/patches/


Rik -- the flu hits, the flu hits, the flu hits -- MORE
| Linux memory management tour guide. H.H.vanRiel@phys.uu.nl |
| Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ |

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/