<tt><font size=2>> From: Tim Bell <Tim.Bell@cern.ch></font></tt>
<br><tt><font size=2>> ...</font></tt>
<br><tt><font size=2>> Discussing with various people in the community,
there seems to be <br>
> interest in a way to</font></tt>
<br><tt><font size=2>> </font></tt>
<br><tt><font size=2>> - Identify
when a hypervisor is being drained or is down <br>
> and inventory its VMs</font></tt>
<br><tt><font size=2>> - Find the
best practise way of restarting that VM for <br>
> hypervisors still available</font></tt>
<br><tt><font size=2>> o Live migration</font></tt>
<br><tt><font size=2>> o Cold migration</font></tt>
<br><tt><font size=2>> - Defining
policies for the remaining cases</font></tt>
<br><tt><font size=2>> o Restart from base image</font></tt>
<br><tt><font size=2>> o Suspend</font></tt>
<br><tt><font size=2>> o Delete</font></tt>
<br><tt><font size=2>> </font></tt>
<br><tt><font size=2>> This touches multiple components from Nova/Cinder/Quantum
(at minimum).</font></tt>
<br><tt><font size=2>> </font></tt>
<br><tt><font size=2>> It also touches some cloud architecture questions
if OpenStack can <br>
> start to move into the low hanging fruit parts of service consolidation.</font></tt>
<br><tt><font size=2>> </font></tt>
<br><tt><font size=2>> I’d like to have some form of summit discussion
in Hong Kong around <br>
> these topics but it is not clear where it fits.</font></tt>
<br><tt><font size=2>> </font></tt>
<br><tt><font size=2>> Are there others who feel similarly ? How can
we fit it in ?</font></tt>
<br>
<br><tt><font size=2>When there are multiple viable choices, I think direction
should be taken from higher layers. The operation of draining a hypervisor
can be parameterized, the VMs themselves can be tagged, by an indication
of which to do.</font></tt>
<br>
<br><tt><font size=2>I myself am working primarily on holistic infrastructure
scheduling, which includes quiescing and draining hypervisors among the
things it can do. Holistic scheduling works under the direction of
a template/pattern/topology that describes a set of interacting resources
and their relationships, and so is able to make a good decision about where
VMs should move to.</font></tt>
<br>
<br><tt><font size=2>Re-starting a VM can require software coordination.</font></tt>
<br>
<br><tt><font size=2>I think holistic infrastructure scheduling is logically
downstream from software coordination and upstream from infrastructure
orchestration. I think the ambitions for Heat are expanding to include
the latter two, and so must also have something to do with holistic infrastructure
scheduling.</font></tt>
<br>
<br><tt><font size=2>Regards,</font></tt>
<br><tt><font size=2>Mike</font></tt>