[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20110614012116.GH5341@mgebm.net>
Date: Mon, 13 Jun 2011 21:21:16 -0400
From: Eric B Munson <emunson@...bm.net>
To: Glauber Costa <glommer@...hat.com>
Cc: kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
Rik van Riel <riel@...hat.com>,
Jeremy Fitzhardinge <jeremy.fitzhardinge@...rix.com>,
Peter Zijlstra <peterz@...radead.org>,
Avi Kivity <avi@...hat.com>,
Anthony Liguori <aliguori@...ibm.com>
Subject: Re: [PATCH 6/7] KVM-GST: adjust scheduler cpu power
On Mon, 13 Jun 2011, Glauber Costa wrote:
> This is a first proposal for using steal time information
> to influence the scheduler. There are a lot of optimizations
> and fine grained adjustments to be done, but it is working reasonably
> so far for me (mostly)
>
> With this patch (and some host pinnings to demonstrate the situation),
> two vcpus with very different steal time (Say 80 % vs 1 %) will not get
> an even distribution of processes. This is a situation that can naturally
> arise, specially in overcommited scenarios. Previosly, the guest scheduler
> would wrongly think that all cpus have the same ability to run processes,
> lowering the overall throughput.
>
> Signed-off-by: Glauber Costa <glommer@...hat.com>
> CC: Rik van Riel <riel@...hat.com>
> CC: Jeremy Fitzhardinge <jeremy.fitzhardinge@...rix.com>
> CC: Peter Zijlstra <peterz@...radead.org>
> CC: Avi Kivity <avi@...hat.com>
> CC: Anthony Liguori <aliguori@...ibm.com>
> CC: Eric B Munson <emunson@...bm.net>
Tested-by: Eric B Munson <emunson@...bm.net>
Download attachment "signature.asc" of type "application/pgp-signature" (491 bytes)
Powered by blists - more mailing lists