<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: More on VMware ESX NIC Utilization</title>
	<atom:link href="http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/</link>
	<description>The weblog of an IT pro specializing in virtualization, storage, and servers</description>
	<lastBuildDate>Wed, 08 Feb 2012 17:13:47 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=</generator>
	<item>
		<title>By: David</title>
		<link>http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/comment-page-1/#comment-43423</link>
		<dc:creator>David</dc:creator>
		<pubDate>Wed, 21 Jan 2009 03:38:52 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/#comment-43423</guid>
		<description>Hey Scott,
I know that you posted this a few months ago, but just wanted to share our observations.  We&#039;ve actually had better and more even distribution using &quot;Route based on MAC hash&quot;.  Technically, both MAC hash and Port ID should deliver the same result...the determination is simply based on different factors, but Port ID seemed to bias NICs higher in the failover order too much, like you mentioned.

Also, there&#039;s a Cisco document out (published May 2008) that talks about this.  I use it as my &quot;bible&quot; for ESX network config design in a Cisco environment.  It can be found here:

http://www.cisco.com/en/US/docs/solutions/Enterprise/Data_Center/vmware/VMwaredg.pdf

Hope this helps someone out there...

- David</description>
		<content:encoded><![CDATA[<p>Hey Scott,<br />
I know that you posted this a few months ago, but just wanted to share our observations.  We&#8217;ve actually had better and more even distribution using &#8220;Route based on MAC hash&#8221;.  Technically, both MAC hash and Port ID should deliver the same result&#8230;the determination is simply based on different factors, but Port ID seemed to bias NICs higher in the failover order too much, like you mentioned.</p>
<p>Also, there&#8217;s a Cisco document out (published May 2008) that talks about this.  I use it as my &#8220;bible&#8221; for ESX network config design in a Cisco environment.  It can be found here:</p>
<p><a href="http://www.cisco.com/en/US/docs/solutions/Enterprise/Data_Center/vmware/VMwaredg.pdf" rel="nofollow">http://www.cisco.com/en/US/docs/solutions/Enterprise/Data_Center/vmware/VMwaredg.pdf</a></p>
<p>Hope this helps someone out there&#8230;</p>
<p>- David</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ook oude netwerk-artikelen over ESX nog actueel &#171; EarlyBert</title>
		<link>http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/comment-page-1/#comment-42895</link>
		<dc:creator>Ook oude netwerk-artikelen over ESX nog actueel &#171; EarlyBert</dc:creator>
		<pubDate>Sun, 21 Dec 2008 15:35:50 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/#comment-42895</guid>
		<description>[...] Twee artikelen over load balancing in de virtuele switch van ESX (hier en hier). [...]</description>
		<content:encoded><![CDATA[<p>[...] Twee artikelen over load balancing in de virtuele switch van ESX (hier en hier). [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: John</title>
		<link>http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/comment-page-1/#comment-41940</link>
		<dc:creator>John</dc:creator>
		<pubDate>Wed, 15 Oct 2008 04:05:11 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/#comment-41940</guid>
		<description>Hi Scott,
We have an ESX box with two pNIC&#039;s teamed together and connected to a layer 2 3750 switch in an Etherchannel.  Problem is when we change the vSwitch to use IP hash the switch looses the plot regarding VLANs and skipping between ports.

Will this only work on a layer 3 version of the switch?  Thanks John</description>
		<content:encoded><![CDATA[<p>Hi Scott,<br />
We have an ESX box with two pNIC&#8217;s teamed together and connected to a layer 2 3750 switch in an Etherchannel.  Problem is when we change the vSwitch to use IP hash the switch looses the plot regarding VLANs and skipping between ports.</p>
<p>Will this only work on a layer 3 version of the switch?  Thanks John</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: fred</title>
		<link>http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/comment-page-1/#comment-41909</link>
		<dc:creator>fred</dc:creator>
		<pubDate>Sat, 11 Oct 2008 07:10:21 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/#comment-41909</guid>
		<description>Thaks Scott, I&#039;m happy to see that i&#039;m not alone: it&#039;s totally true for the originating port id ! I&#039;ve made some tests with iperf and i saw the same thing : severals vm using the same physical nic !
Just a question about the lag: what is the maximum bandwith a vm can have with two physical nics used with ip hash? 1 Gbits or 2 ?
Thks !</description>
		<content:encoded><![CDATA[<p>Thaks Scott, I&#8217;m happy to see that i&#8217;m not alone: it&#8217;s totally true for the originating port id ! I&#8217;ve made some tests with iperf and i saw the same thing : severals vm using the same physical nic !<br />
Just a question about the lag: what is the maximum bandwith a vm can have with two physical nics used with ip hash? 1 Gbits or 2 ?<br />
Thks !</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Tom</title>
		<link>http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/comment-page-1/#comment-41900</link>
		<dc:creator>Tom</dc:creator>
		<pubDate>Fri, 10 Oct 2008 14:34:45 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/#comment-41900</guid>
		<description>Thanks Scott, it is always nice to have a quick and easy reference for this type of information.</description>
		<content:encoded><![CDATA[<p>Thanks Scott, it is always nice to have a quick and easy reference for this type of information.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Virgil</title>
		<link>http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/comment-page-1/#comment-41887</link>
		<dc:creator>Virgil</dc:creator>
		<pubDate>Wed, 08 Oct 2008 23:55:11 +0000</pubDate>
		<guid isPermaLink="false">http://blog.scottlowe.org/2008/10/08/more-on-vmware-esx-nic-utilization/#comment-41887</guid>
		<description>Scott, thanks for the refresh.  Regarding NIC teaming inside the VM, I&#039;ve been thinking (and assuming) the same thing.

Will get back you when I get a chance to test it.

Virgil</description>
		<content:encoded><![CDATA[<p>Scott, thanks for the refresh.  Regarding NIC teaming inside the VM, I&#8217;ve been thinking (and assuming) the same thing.</p>
<p>Will get back you when I get a chance to test it.</p>
<p>Virgil</p>
]]></content:encoded>
	</item>
</channel>
</rss>

