<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:clearspace="http://www.jivesoftware.com/xmlns/clearspace/rss" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:wfw="http://wellformedweb.org/CommentAPI/" xmlns:opensearch="http://a9.com/-/spec/opensearch/1.1/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>Performance &amp; VMmark : Co-scheduling SMP VMs in VMware ESX Server : Comments</title>
    <link>http://communities.vmware.com/docs/DOC-4960#comments</link>
    <description>Comments on : Co-scheduling SMP VMs in VMware ESX Server</description>
    <language>en</language>
    <pubDate>Thu, 22 May 2008 03:39:49 GMT</pubDate>
    <generator>Clearspace 1.10.12 (http://jivesoftware.com/products/clearspace/)</generator>
    <dc:date>2008-05-22T03:39:49Z</dc:date>
    <dc:language>en</dc:language>
    <item>
      <title>RE: Co-scheduling SMP VMs in VMware ESX Server</title>
      <link>http://communities.vmware.com/docs/DOC-4960#comments-6265</link>
      <description>A question that gets raised fairly frequently is "is it better to put vSMP VMs together on a host, or mix them with single VCPU VMs?" Based on the article above, I submit that - as long as you're not accumulating significant skew, it shouldn't make much difference.&lt;br /&gt;
&lt;br /&gt;
Could you address how NUMA impacts CPU scheduling? And also, I've heard that pinning a vCPU to a pCore has a negative impact on the scheduler...can you comment?&lt;br /&gt;
&lt;br /&gt;
Thanks!&lt;br /&gt;
&lt;br /&gt;
Ken Cline&lt;br /&gt;
Technical Director, Virtualization&lt;br /&gt;
&lt;a class="jive-link-external" href="http://www.wellslanders.com"&gt;Wells Landers&lt;/a&gt;&lt;br /&gt;
VMware Communities User Moderator</description>
      <pubDate>Thu, 22 May 2008 03:39:49 GMT</pubDate>
      <author>Ken.Cline</author>
      <guid>http://communities.vmware.com/docs/DOC-4960#comments-6265</guid>
      <dc:date>2008-05-22T03:39:49Z</dc:date>
      <clearspace:dateToText>1 year, 6 months ago</clearspace:dateToText>
    </item>
    <item>
      <title>RE: Co-scheduling SMP VMs in VMware ESX Server</title>
      <link>http://communities.vmware.com/docs/DOC-4960#comments-6445</link>
      <description>Hi, Ken,&lt;br /&gt;
&lt;br /&gt;
I created a broader scheduling topic called "VMkernel Scheduler" that answers some of your questions.  Since your questions are strictly co-scheduling questions, I figured that to be a better place to provide answers.&lt;br /&gt;
&lt;br /&gt;
I haven't yet provided the supporting data for these claims but hope to do so by the time we've finished seeding content.  That should take a couple of months.  In the mean time, please trust me that I'm not lying on the claims I've made.  &lt;img class="jive-emoticon" border="0" src="http://communities.vmware.com/images/emoticons/happy.gif" alt=":)" /&gt;&lt;br /&gt;
&lt;br /&gt;
Scott</description>
      <pubDate>Tue, 27 May 2008 23:05:45 GMT</pubDate>
      <author>drummonds</author>
      <guid>http://communities.vmware.com/docs/DOC-4960#comments-6445</guid>
      <dc:date>2008-05-27T23:05:45Z</dc:date>
      <clearspace:dateToText>1 year, 6 months ago</clearspace:dateToText>
    </item>
    <item>
      <title>RE: Co-scheduling SMP VMs in VMware ESX Server</title>
      <link>http://communities.vmware.com/docs/DOC-4960#comments-6491</link>
      <description>Thanks Scott, I'll go read the other doc, too!&lt;br /&gt;
&lt;br /&gt;
Ken Cline&lt;br /&gt;
Technical Director, Virtualization&lt;br /&gt;
&lt;a class="jive-link-external" href="http://www.wellslanders.com"&gt;Wells Landers&lt;/a&gt;&lt;br /&gt;
VMware Communities User Moderator</description>
      <pubDate>Thu, 29 May 2008 13:04:40 GMT</pubDate>
      <author>Ken.Cline</author>
      <guid>http://communities.vmware.com/docs/DOC-4960#comments-6491</guid>
      <dc:date>2008-05-29T13:04:40Z</dc:date>
      <clearspace:dateToText>1 year, 5 months ago</clearspace:dateToText>
    </item>
    <item>
      <title>RE: Co-scheduling SMP VMs in VMware ESX Server</title>
      <link>http://communities.vmware.com/docs/DOC-4960#comments-7256</link>
      <description>Scott, it's e1 from Singapore (we met in Bangkok during the QBR).&lt;br /&gt;
&lt;br /&gt;
When we relax it in 3.x, do you mean starting from 3.0, or a certain update of 3.0 (e.g. 3.0.2)?&lt;br /&gt;
&lt;br /&gt;
In 3.5.x, did we relax it further? If yes, could you provide more details?&lt;br /&gt;
&lt;br /&gt;
Will email you separately which customer asked me the question.&lt;br /&gt;
&lt;br /&gt;
Tx&lt;br /&gt;
e1</description>
      <pubDate>Sat, 19 Jul 2008 00:30:22 GMT</pubDate>
      <author>Iwan Rahabok</author>
      <guid>http://communities.vmware.com/docs/DOC-4960#comments-7256</guid>
      <dc:date>2008-07-19T00:30:22Z</dc:date>
      <clearspace:dateToText>1 year, 4 months ago</clearspace:dateToText>
    </item>
    <item>
      <title>RE: Co-scheduling SMP VMs in VMware ESX Server</title>
      <link>http://communities.vmware.com/docs/DOC-4960#comments-11200</link>
      <description>If we enable Hyper Threading, how vSMP work?&lt;br /&gt;
I heard that LCPU can be HEC(Hardware Execution Context?).&lt;br /&gt;
Is that right?&lt;br /&gt;
We can see it in the text I&amp;#38;C course from VMware.&lt;br /&gt;
If so, LCU fix this issue?&lt;br /&gt;
For instance,&lt;br /&gt;
There is only 2 LCPC(not PCPU)on 1 PCPU, 2vSMP VM can run with no waiting?</description>
      <pubDate>Wed, 15 Apr 2009 23:46:29 GMT</pubDate>
      <author>man33taro</author>
      <guid>http://communities.vmware.com/docs/DOC-4960#comments-11200</guid>
      <dc:date>2009-04-15T23:46:29Z</dc:date>
      <clearspace:dateToText>7 months, 1 week ago</clearspace:dateToText>
    </item>
  </channel>
</rss>

