[openstack-dev] [POC] Introduce an auto-converge policy to speedup migration
fanc.fnst at cn.fujitsu.com
Tue May 16 06:41:03 UTC 2017
Sorry for no Cc to you, I remember I have added cc
On Mon, May 15, 2017 at 01:30:39PM +0800, Chao Fan wrote:
>On Thu, May 11, 2017 at 02:34:16PM -0400, Chris Friesen wrote:
>>On 05/11/2017 05:58 AM, Chao Fan wrote:
>>> Hi all,
>>> We plan to develop a policy about auto-converge, which can set cpu
>>> throttle value automatically according to the workload
>>> (dirty-pages-rate). It uses the API of libvirt to set the
>>> cpu-throttle-initial and cpu-throttle-increment.
>>> But the spec file of nova shows the dependent API is not accepted
>>> by OpenStack:
>>> The initial decrease and increment size can be adjusted during
>>> the live migration process via the libvirt API. However these API calls
>>> are experimental so nova will not be using them.
>>> So I am wondering if OpenStack is willing to use this API and accept
>>> the policy mentioned above.
>>Just to clarify, as I understand it:
>>1) You are pointing out that the auto-live-migration spec from Newton says
>>that the libvirt APIs to set the initial and increment throttle values are
>>experimental and thus won't be using them.
>Thank you for your reply, and really sorry for delay. Cause I did not
>notice this mail without Cc me.
>The spec file is cloned from https://github.com/openstack/nova-specs.git.
>It looks same to your link.
>>2) You are asking whether these APIs are now stable enough to be used in
>>nova, since you want to propose some mechanism to allow them to be changed.
>>Is that accurate?
>Yes, your understanding is right.
>>OpenStack Development Mailing List (not for usage questions)
>>Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
>OpenStack Development Mailing List (not for usage questions)
>Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
More information about the OpenStack-dev