<div dir="ltr"><div>Donald,<br><br></div>By "selection", I think Jenny means identifying whether and which active VM should be migrated, once the current Nova scheduler only deals with the VM in the momment of its creation or with a specific user input.<br>
</div><div class="gmail_extra"><br><br><div class="gmail_quote">2014-04-24 12:08 GMT-03:00 Dugger, Donald D <span dir="ltr"><<a href="mailto:donald.d.dugger@intel.com" target="_blank">donald.d.dugger@intel.com</a>></span>:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div link="blue" vlink="purple" lang="EN-US">
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">Jenny-<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">You should look at the `Propose Scheduler Library blueprint’:<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">
<a href="https://review.openstack.org/#/c/82133/9" target="_blank">https://review.openstack.org/#/c/82133/9</a><u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">This BP is to create a client library for making calls to the scheduler. If you base your work upon this library then you shouldn’t need to care about whether
the Core Scheduler is the Nova integrated scheduler or the Gantt separated scheduler, the library will call `a` scheduler as appropriate.<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">Having said that, I’m not sure I understand the distinction you are seeing between `selection’ and `placement’. The current scheduler filters all hosts based
upon filters (the selection part) and then the weighting function finds the best node to host the VM (the placement part). Seems to me the current scheduler does both of those tasks. (We can argue the effectiveness/efficiency of the current implementation
but I think it’s functionally complete.)<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">Also, have you proposed a session at the Juno summit on your proposal for dynamic scheduling, seems like it would be appropriate.<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">--<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">Don Dugger<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">"Censeo Toto nos in Kansa esse decisse." - D. Gale<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d">Ph: 303/443-3786<u></u><u></u></span></p>
</div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<div>
<div style="border:none;border-top:solid #e1e1e1 1.0pt;padding:3.0pt 0in 0in 0in">
<p class="MsoNormal"><b><span style="font-size:11.0pt;font-family:"Calibri","sans-serif"">From:</span></b><span style="font-size:11.0pt;font-family:"Calibri","sans-serif""> Jiangying (Jenny) [mailto:<a href="mailto:jenny.jiangying@huawei.com" target="_blank">jenny.jiangying@huawei.com</a>]
<br>
<b>Sent:</b> Thursday, April 24, 2014 3:36 AM<br>
<b>To:</b> OpenStack Development Mailing List (not for usage questions)<br>
<b>Subject:</b> Re: [openstack-dev] [nova][ceilometer][gantt] Dynamic scheduling<u></u><u></u></span></p>
</div>
</div><div><div class="h5">
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">Hi,<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">We have checked that gantt now just made a synced up copy of the code in nova.
<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">We still think dynamic scheduling will be a benefit of the nova scheduler (or gantt later). The main difference between static and dynamic scheduling
is that static scheduling is a vm placement problem, while dynamic scheduling deals with both vm selection and vm placement.<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif""><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">Our scheduling mechanism consists of three parts:<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">1. Controller, which triggers the scheduling;<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">2. Data Collector, which collects the resource usage and topo for scheduling;<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">3. Core Scheduler, which decides how to schedule the vms;
<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif""><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">We prefer to reuse the nova scheduler as the Core Scheduler, in order to avoid the possible inconsistent between static scheduling and dynamic
scheduling. The vm selection function will be added into nova scheduler. For Data Collector, we expect to get the performance data from ceilometer and topo from nova.
<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif""><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">There is still one question that where the controller should be implemented?
<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">We regard implementing the controller in nova scheduler as the first choice. And we also consider extending ceilometer.(Ie. When ceilometer discovers
an overload host, an alarm can be reported and it can trigger a vm evacuate.)<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif""><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">Do you have any comments?<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif""><u></u> <u></u></span></p>
<p class="MsoNormal"><span style="font-size:14.0pt;font-family:"Calibri","sans-serif"">Jenny<u></u><u></u></span></p>
<p class="MsoNormal"><span style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<p class="MsoNormal"><b><span style="font-size:10.0pt;font-family:SimSun" lang="ZH-CN">发件人</span></b><b><span style="font-size:10.0pt;font-family:SimSun">:</span></b><span style="font-size:10.0pt;font-family:SimSun">
Henrique Truta [<a href="mailto:henriquecostatruta@gmail.com" target="_blank">mailto:henriquecostatruta@gmail.com</a>]
<br>
<b><span lang="ZH-CN">发送时间</span>:</b> 2014<span lang="ZH-CN">年</span>4<span lang="ZH-CN">月</span>12<span lang="ZH-CN">日</span> 1:00<br>
<b><span lang="ZH-CN">收件人</span>:</b> OpenStack Development Mailing List (not for usage questions)<br>
<b><span lang="ZH-CN">主题</span>:</b> Re: [openstack-dev] [nova] Dynamic scheduling<u></u><u></u></span></p>
<p class="MsoNormal"><span><u></u> <u></u></span></p>
<div>
<p class="MsoNormal"><span>Is there anyone currently working on Neat/Gantt projects? I'd like to contribute to them, as well.<u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span><u></u> <u></u></span></p>
<div>
<p class="MsoNormal"><span>2014-04-11 11:37 GMT-03:00 Andrew Laski <<a href="mailto:andrew.laski@rackspace.com" target="_blank">andrew.laski@rackspace.com</a>>:<u></u><u></u></span></p>
<div>
<p class="MsoNormal"><span>On 04/10/14 at 11:33pm, Oleg Gelbukh wrote:<u></u><u></u></span></p>
<p class="MsoNormal"><span>Andrew,<br>
<br>
Thank you for clarification!<br>
<br>
<br>
On Thu, Apr 10, 2014 at 3:47 PM, Andrew Laski <<a href="mailto:andrew.laski@rackspace.com" target="_blank">andrew.laski@rackspace.com</a>>wrote:<u></u><u></u></span></p>
<p class="MsoNormal"><span><br>
<br>
The scheduler as it currently exists is a placement engine. There is<br>
sufficient complexity in the scheduler with just that responsibility so I<br>
would prefer to see anything that's making runtime decisions separated out.<br>
Perhaps it could just be another service within the scheduler project once<br>
it's broken out, but I think it will be beneficial to have a clear<br>
distinction between placement decisions and runtime monitoring.<u></u><u></u></span></p>
<p class="MsoNormal"><span><br>
<br>
Do you think that auto-scaling could be considered another facet of this<br>
'runtime monitoring' functionality? Now it is a combination of Heat and<br>
Ceilometer. Does it worth moving to hypothetical runtime mobility service<br>
as well?<u></u><u></u></span></p>
<p class="MsoNormal"><span><u></u> <u></u></span></p>
</div>
<p class="MsoNormal"><span>Auto-scaling is certainly a facet of runtime monitoring. But auto-scaling performs actions based on a set of user defined rules and is very visible while the enhancements proposed below are intended
to benefit deployers and be very invisible to users. So the set of allowable actions is very constrained compared to what auto-scaling can do. <br>
In my opinion what's being proposed doesn't seem to fit cleanly into any existing service, so perhaps it could start as a standalone entity. Then once there's something that can be used and demoed a proper place might suggest itself, or it might make sense
to keep it separate.<u></u><u></u></span></p>
<div>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span><u></u> <u></u></span></p>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span><br>
--<br>
Best regards,<br>
Oleg Gelbukh<u></u><u></u></span></p>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span><u></u> <u></u></span></p>
<p class="MsoNormal"><span>--<br>
Best regards,<br>
Oleg Gelbukh<br>
<br>
<br>
On Wed, Apr 9, 2014 at 7:47 PM, Jay Lau <<a href="mailto:jay.lau.513@gmail.com" target="_blank">jay.lau.513@gmail.com</a>> wrote:<br>
<br>
@Oleg, Till now, I'm not sure the target of Gantt, is it for initial<u></u><u></u></span></p>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span>placement policy or run time policy or both, can you help clarify?<br>
<br>
@Henrique, not sure if you know IBM PRS (Platform Resource Scheduler)<br>
[1],<br>
we have finished the "dynamic scheduler" in our Icehouse version (PRS<br>
2.2),<br>
it has exactly the same feature as your described, we are planning a live<br>
demo for this feature in Atlanta Summit. I'm also writing some document<br>
for<br>
run time policy which will cover more run time policies for OpenStack,<br>
but<br>
not finished yet. (My shame for the slow progress). The related blueprint<br>
is [2], you can also get some discussion from [3]<br>
<br>
[1]<br>
<a href="http://www-01.ibm.com/common/ssi/cgi-bin/ssialias?infotype=" target="_blank">http://www-01.ibm.com/common/ssi/cgi-bin/ssialias?infotype=</a><br>
AN&subtype=CA&htmlfid=897/ENUS213-590&appname=USN<br>
[2]<br>
<a href="https://blueprints.launchpad.net/nova/+spec/resource-" target="_blank">https://blueprints.launchpad.net/nova/+spec/resource-</a><br>
optimization-service<br>
[3] <a href="http://markmail.org/~jaylau/OpenStack-DRS" target="_blank">http://markmail.org/~jaylau/OpenStack-DRS</a><br>
<br>
Thanks.<br>
<br>
<br>
2014-04-09 23:21 GMT+08:00 Oleg Gelbukh <<a href="mailto:ogelbukh@mirantis.com" target="_blank">ogelbukh@mirantis.com</a>>:<br>
<br>
Henrique,<u></u><u></u></span></p>
<p class="MsoNormal"><span><br>
You should check out Gantt project [1], it could be exactly the place to<br>
implement such features. It is a generic cross-project Scheduler as a<br>
Service forked from Nova recently.<br>
<br>
[1] <a href="https://github.com/openstack/gantt" target="_blank">https://github.com/openstack/gantt</a><br>
<br>
--<br>
Best regards,<br>
Oleg Gelbukh<br>
Mirantis Labs<br>
<br>
<br>
On Wed, Apr 9, 2014 at 6:41 PM, Henrique Truta <<br>
<a href="mailto:henriquecostatruta@gmail.com" target="_blank">henriquecostatruta@gmail.com</a>> wrote:<br>
<br>
Hello, everyone!<u></u><u></u></span></p>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span><br>
I am currently a graduate student and member of a group of contributors<br>
to OpenStack. We believe that a dynamic scheduler could improve the<br>
efficiency of an OpenStack cloud, either by rebalancing nodes to<br>
maximize<br>
performance or to minimize the number of active hosts, in order to<br>
minimize<br>
energy costs. Therefore, we would like to propose a dynamic scheduling<br>
mechanism to Nova. The main idea is using the Ceilometer information<br>
(e.g.<br>
RAM, CPU, disk usage) through the ceilometer-client and dinamically<br>
decide<br>
whether a instance should be live migrated.<br>
<br>
This might me done as a Nova periodic task, which will be executed<br>
every<br>
once in a given period or as a new independent project. In both cases,<br>
the<br>
current Nova scheduler will not be affected, since this new scheduler<br>
will<br>
be pluggable. We have done a search and found no such initiative in the<br>
OpenStack BPs. Outside the community, we found only a recent IBM<br>
announcement for a similiar feature in one of its cloud products.<br>
<br>
A possible flow is: In the new scheduler, we periodically make a call<br>
to<br>
Nova, get the instance list from a specific host and, for each<br>
instance, we<br>
make a call to the ceilometer-client (e.g. $ ceilometer statistics -m<br>
cpu_util -q resource=$INSTANCE_ID) and then, according to some specific<br>
parameters configured by the user, analyze the meters and do the proper<br>
migrations.<br>
<br>
Do you have any comments or suggestions?<br>
<br>
--<br>
Ítalo Henrique Costa Truta<br>
<br>
<br>
<br>
_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org" target="_blank">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br>
<u></u><u></u></span></p>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span>_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org" target="_blank">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br>
<u></u><u></u></span></p>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span><br>
--<br>
Thanks,<br>
<br>
Jay<br>
<br>
_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org" target="_blank">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br>
<u></u><u></u></span></p>
<p class="MsoNormal"><span> _______________________________________________<u></u><u></u></span></p>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span>OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org" target="_blank">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><u></u><u></u></span></p>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span><br>
<br>
_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org" target="_blank">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><u></u><u></u></span></p>
<blockquote style="border:none;border-left:solid #cccccc 1.0pt;padding:0in 0in 0in 6.0pt;margin-left:4.8pt;margin-top:5.0pt;margin-right:0in;margin-bottom:5.0pt">
<p class="MsoNormal"><span><u></u> <u></u></span></p>
<p class="MsoNormal"><span>_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org" target="_blank">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><u></u><u></u></span></p>
</blockquote>
<p class="MsoNormal"><span><br>
<br>
_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org" target="_blank">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><u></u><u></u></span></p>
</div>
</div>
</div>
<p class="MsoNormal"><span><br>
<br clear="all">
<br>
-- <u></u><u></u></span></p>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span style="color:#999999">--<br>
Ítalo Henrique Costa Truta</span><span><u></u><u></u></span></p>
</div>
</div>
</div></div></div>
</div>
<br>_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br></blockquote></div><br><br clear="all"><br>-- <br><div dir="ltr"><font color="#999999">--<br>Ítalo Henrique Costa Truta<br><br></font><br></div>
</div>