Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 14 Sep 2006 19:43:27 +0200
From:      Pieter de Goeje <pieter@degoeje.nl>
To:        Gary Corcoran <gcorcoran@rcn.com>
Cc:        freebsd-hackers@freebsd.org
Subject:   Re: numbers don't lie ...
Message-ID:  <200609141943.28218.pieter@degoeje.nl>
In-Reply-To: <45099123.4000500@rcn.com>
References:  <E1GNOLq-000DC2-1Q@cs1.cs.huji.ac.il> <863bauk3gp.fsf@dwp.des.no> <45099123.4000500@rcn.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thursday 14 September 2006 19:28, Gary Corcoran wrote:
> The confusing thing is that I thought 'real' time should be >= 'user' +
> 'sys'. But here 'user' is much greater than 'real' for both machines!  The
> sense I got from the other messages in this thread is that 'user' time is
> somewhat meaningless (i.e. unreliable as a measure) in a multi-CPU and/or
> hyperthreading environment.  Can you clarify?

user time = time spent in userland on all logical processors combined.

The right equation is: real * ncpus > user + sys, where ncpus = number of 
active logical processors. 
In the optimal case (perfect parallelism): real * ncpus = user + sys


- Pieter



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200609141943.28218.pieter>