Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 29 Dec 2005 10:48:16 -0500
From:      Martin Cracauer <cracauer@cons.org>
To:        Kiffin Gish <kiffin.gish@planet.nl>
Cc:        freebsd-questions@freebsd.org
Subject:   Re: Using SCHED_ULE vs. SCHED_4BSD ...
Message-ID:  <20051229104816.B581@cons.org>
In-Reply-To: <1135855260.1034.5.camel@localhost>; from kiffin.gish@planet.nl on Thu, Dec 29, 2005 at 12:21:00PM %2B0100
References:  <1135855260.1034.5.camel@localhost>

next in thread | previous in thread | raw e-mail | index | archive | help
Kiffin Gish wrote on Thu, Dec 29, 2005 at 12:21:00PM +0100: 
> I recently upgraded from 5.4 to 6.0 and noticed the introduction of
> option SCHED_ULE for supporting multi-processor environments.
> 
> However, I understood that using SCHED_ULE with only one CPU can also
> improve performance significantly.
> 
> Is this true, and if so, what are the risks involved dropping good old
> SCHED_4BSD for the new-and-improved scheduler?

I have not even noticed an increase in performance from ULE when
running my benchmark suite on a two-processor system.

Some people say ULE has problems, but mostly on high CPU counts.

So it probably doesn't matter either way.

Martin
-- 
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Martin Cracauer <cracauer@cons.org>   http://www.cons.org/cracauer/
FreeBSD - where you want to go, today.      http://www.freebsd.org/



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20051229104816.B581>