<div dir="ltr">Jiang,<div><br></div><div>I would like to reduce "magic"</div><div><br></div><div>1) We are using already RPC (because all compute nodes update are done in DB via conductor (which means RPC call). </div>
<div>So count of RPC calls and size of msg will be the same. </div><div><br></div><div>2) There is no lazy load when you have to fetch all data about all compute nodes on every request to scheduler. </div><div><br></div><div>
3) Object models are off topic </div><div><br></div><div>Best regards,</div><div>Boris Pavlovic</div><div><br></div><div>Mirantis Inc. </div><div><br></div><div><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">
On Fri, Jul 19, 2013 at 11:23 PM, Jiang, Yunhong <span dir="ltr"><<a href="mailto:yunhong.jiang@intel.com" target="_blank">yunhong.jiang@intel.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div lang="ZH-CN" link="blue" vlink="purple">
<div>
<p class="MsoNormal"><span lang="EN-US">Boris<u></u><u></u></span></p>
<p class="MsoNormal"><span lang="EN-US"> I think you in fact covered two topic, one is if use db or rpc for communication. This has been discussed a lot. But I didn’t find the conclusion. From the discussion, seems the key thing is the fan out messages.
I’d suggest you to bring this to scheduler sub meeting.<u></u><u></u></span></p>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="MsoNormal"><span lang="EN-US"><a href="http://eavesdrop.openstack.org/meetings/scheduler/2013/scheduler.2013-06-11-14.59.log.html" target="_blank">http://eavesdrop.openstack.org/meetings/scheduler/2013/scheduler.2013-06-11-14.59.log.html</a><u></u><u></u></span></p>
<p class="MsoNormal"><span lang="EN-US"><a href="http://www.mail-archive.com/openstack-dev@lists.openstack.org/msg00070.html" target="_blank">http://www.mail-archive.com/openstack-dev@lists.openstack.org/msg00070.html</a><u></u><u></u></span></p>
<p class="MsoNormal"><span lang="EN-US"><a href="http://comments.gmane.org/gmane.comp.cloud.openstack.devel/23" target="_blank">http://comments.gmane.org/gmane.comp.cloud.openstack.devel/23</a>
<u></u><u></u></span></p>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="MsoNormal"><span lang="EN-US"> The second topic is adding extra tables to compute nodes. I think we need the lazy loading for the compute node, and also I think with object model, we can further improve it if we utilize the compute node object.<u></u><u></u></span></p>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="MsoNormal"><span lang="EN-US">Thanks<u></u><u></u></span></p>
<p class="MsoNormal"><span lang="EN-US">--jyh<u></u><u></u></span></p>
<p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<p class="MsoNormal"><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri","sans-serif";color:#1f497d"><u></u> <u></u></span></p>
<div style="border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt">
<div>
<div style="border:none;border-top:solid #b5c4df 1.0pt;padding:3.0pt 0cm 0cm 0cm">
<p class="MsoNormal"><b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">From:</span></b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif""> Boris Pavlovic [mailto:<a href="mailto:boris@pavlovic.me" target="_blank">boris@pavlovic.me</a>]
<br>
<b>Sent:</b> Friday, July 19, 2013 10:07 AM</span></p><div class="im"><br>
<b>To:</b> OpenStack Development Mailing List<br>
<b>Subject:</b> Re: [openstack-dev] [Nova] New DB column or new DB table?<u></u><u></u></div><p></p>
</div>
</div>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
<div>
<p class="MsoNormal"><span lang="EN-US">Hi all, <u></u><u></u></span></p><div><div class="h5">
<div>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US">We have to much different branches about scheduler (so I have to repeat here also).<u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US">I am against to add some extra tables that will be joined to compute_nodes table on each scheduler request (or adding large text columns).<u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US">Because it make our non scalable scheduler even less scalable. <u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US">Also if we just remove DB between scheduler and compute nodes we will get really good improvement in all aspects (performance, db load, network traffic, scalability )<u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US">And also it will be easily to use another resources provider (cinder, ceilometer e.g..) in Nova scheduler. <u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US">And one more thing this all could be really simple implement in current Nova, without big changes <u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US"> <a href="https://docs.google.com/document/d/1_DRv7it_mwalEZzLy5WO92TJcummpmWL4NWsWf0UWiQ/edit?usp=sharing" target="_blank">https://docs.google.com/document/d/1_DRv7it_mwalEZzLy5WO92TJcummpmWL4NWsWf0UWiQ/edit?usp=sharing</a><u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US">Best regards,<u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US">Boris Pavlovic <u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US"><u></u> <u></u></span></p>
</div>
<div>
<p class="MsoNormal"><span lang="EN-US">Mirantis Inc.<u></u><u></u></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span lang="EN-US"><u></u> <u></u></span></p>
<div>
<p class="MsoNormal"><span lang="EN-US">On Fri, Jul 19, 2013 at 8:44 PM, Dan Smith <<a href="mailto:dms@danplanet.com" target="_blank">dms@danplanet.com</a>> wrote:<u></u><u></u></span></p>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span lang="EN-US">> IIUC, Ceilometer is currently a downstream consumer of data from<br>
> Nova, but no functionality in Nova is a consumer of data from<br>
> Ceilometer. This is good split from a security separation point of<br>
> view, since the security of Nova is self-contained in this<br>
> architecture.<br>
><br>
> If Nova schedular becomes dependant on data from ceilometer, then now<br>
> the security of Nova depends on the security of Ceilometer, expanding<br>
> the attack surface. This is not good architecture IMHO.<u></u><u></u></span></p>
</div>
<p class="MsoNormal"><span lang="EN-US">Agreed.<u></u><u></u></span></p>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><span lang="EN-US"><br>
> At the same time, I hear your concerns about the potential for<br>
> duplication of stats collection functionality between Nova &<br>
> Ceilometer. I don't think we neccessarily need to remove 100% of<br>
> duplication. IMHO probably the key thing is for the virt drivers to<br>
> expose a standard API for exporting the stats, and make sure that<br>
> both ceilometer & nova schedular use the same APIs and ideally the<br>
> same data feed, so we're not invoking the same APIs twice to get the<br>
> same data.<u></u><u></u></span></p>
</div>
<p class="MsoNormal"><span lang="EN-US">I imagine there's quite a bit that could be shared, without dependency<br>
between the two. Interfaces out of the virt drivers may be one, and the<br>
code that boils numbers into useful values, as well as perhaps the<br>
format of the JSON blobs that are getting shoved into the database.<br>
Perhaps a ceilo-core library with some very simple primitives and<br>
definitions could be carved out, which both nova and ceilometer could<br>
import for consistency, without a runtime dependency?<br>
<span style="color:#888888"><br>
--Dan</span><u></u><u></u></span></p>
<div>
<div>
<p class="MsoNormal"><span lang="EN-US"><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 lang="EN-US"><u></u> <u></u></span></p>
</div>
</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></div>