Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 30 Aug 2011 09:24:59 +0100
From:      Bruce Cran <bruce@cran.org.uk>
To:        Volodymyr Kostyrko <c.kworr@gmail.com>
Cc:        "freebsd-performance@freebsd.org" <freebsd-performance@freebsd.org>, freebsd Current <freebsd-current@freebsd.org>, "Hartmann, O." <ohartman@zedat.fu-berlin.de>
Subject:   Re: http://www.freebsd.org/marketing/os-comparison.html
Message-ID:  <4E5C9E5B.6010404@cran.org.uk>
In-Reply-To: <4E5BEF65.2010502@gmail.com>
References:  <4E5941D6.9090106@zedat.fu-berlin.de> <4E5BEF65.2010502@gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On 29/08/2011 20:58, Volodymyr Kostyrko wrote:
> = SMP =
>  * (?something about comparing other shedulers with SCHED_ULE), (?some 
> rt stuff), (?some comparison with other interesting shedulers, like 
> DragonflyBSD and QNX).

 From a recent post to -questions:

"Alas, during a recent kernel build, I used the -j2 command line option 
in "make" and watched as the scheduler repeatedly assigned two instances 
of cc (the most CPU-intensive program) to the same core."

I'm not sure this is something we're really better at, unfortunately: I 
know I've watched Windows really grok multi-socket, multi-core 
HyperThreaded systems and prefer real cores on the same NUMA node when 
running a multi-threaded application, whereas it seems FreeBSD struggles 
sometimes.

-- 
Bruce Cran



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4E5C9E5B.6010404>