alaric at metrocast.net
Fri May 21 08:59:14 CDT 2010
On 05/21/10 08:56, gsm at mendelson.com wrote:
> Up until recently Linux had a scheduler that really performed poorly under
> heavy loads. In the 2.6 kernel (or around that time anyway) a new scheduler
> was implemented that supposedly fix those problems, but I have no way of
> verifying the hype.
I can tell you that this machine, with a single AthlonXP 2400+ CPU and a
2.6.33 kernel, remains usably responsive (except for Firefox, which I
run heavily niced) with a dozen or so simultaneous gcc processes
running. (This typically comes about when I'm configured for distcc,
with make -j13 or -j15, but the ebuild in question disallows distcc.)
And I am NOT running the new low-latency desktop scheduler.
Phil Stracchino, CDK#2 DoD#299792458 ICBM: 43.5607, -71.355
alaric at caerllewys.net alaric at metrocast.net phil at co.ordinate.org
Renaissance Man, Unix ronin, Perl hacker, Free Stater
It's not the years, it's the mileage.
More information about the rescue