|
From: | Anthony Liguori |
Subject: | Re: [Qemu-devel] Re: [PATCH 4/4] Try not to exceed max downtime on stage3 |
Date: | Tue, 12 Jan 2010 09:07:17 -0600 |
User-agent: | Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.5) Gecko/20091209 Fedora/3.0-4.fc12 Lightning/1.0pre Thunderbird/3.0 |
On 01/12/2010 05:51 AM, Jan Kiszka wrote:
Liran Schour wrote:Move to stage3 only when remaining work can be done below max downtime. To make sure the process will converge we will try only MAX_DIRTY_ITERATIONS.OK, that explains now patch 2. But do we have such barrier for memory migration as well?
No, we explicitly don't because making that decision is a management tool job.
A management tool can force convergence by explicitly stopping the guest. Iterations is a bad metric because there's no real useful meaning to a user.Time is probably the best metric and it's easy for a management tool to set a timer to stop the guest if it hasn't completed by a certain time period.
Regards, Anthony Liguori
[Prev in Thread] | Current Thread | [Next in Thread] |