<?xml version="1.0" encoding="UTF-8" ?>
<?xml-stylesheet type="text/xsl" href="http://blogs.msdn.com/utility/FeedStylesheets/rss.xsl" media="screen"?><rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:slash="http://purl.org/rss/1.0/modules/slash/" xmlns:wfw="http://wellformedweb.org/CommentAPI/"><channel><title>Hyper-V: How to make sure you are getting the best performance when doing performance comparisons</title><link>http://blogs.msdn.com/b/tvoellm/archive/2008/04/19/hyper-v-how-to-make-sure-you-are-getting-the-best-performance-when-doing-performance-comparisons.aspx</link><description>As we have moved from Beta to RC the number of questions I have been getting around Hyper-V performance have grown quite a bit. It’s time to put some tips out on how to avoid some common pitfalls. 
 Pitfall 1: The first and most common has been running</description><dc:language>en-US</dc:language><generator>Telligent Evolution Platform Developer Build (Build: 5.6.50428.7875)</generator><item><title>re: Hyper-V: How to make sure you are getting the best performance when doing performance comparisons</title><link>http://blogs.msdn.com/b/tvoellm/archive/2008/04/19/hyper-v-how-to-make-sure-you-are-getting-the-best-performance-when-doing-performance-comparisons.aspx#8994202</link><pubDate>Fri, 10 Oct 2008 15:51:35 GMT</pubDate><guid isPermaLink="false">91d46819-8472-40ad-a661-2c78acb4018c:8994202</guid><dc:creator>William Fields</dc:creator><description>&lt;P&gt;Hello,&lt;/P&gt;
&lt;P&gt;I'd like to see some file transfer performance numbers on your end. Also, I'd like to know (or clarify) a couple things.&lt;/P&gt;
&lt;P&gt;My understanding (&lt;A href="http://blogs.technet.com/jhoward/archive/2008/06/16/how-does-basic-networking-work-in-hyper-v.aspx" target=_new rel=nofollow&gt;http://blogs.technet.com/jhoward/archive/2008/06/16/how-does-basic-networking-work-in-hyper-v.aspx&lt;/A&gt;) is that when transfering data from parent to child through the virtual network switch (Synthetic NIC, and VMBus active), there should be no actual traffic between the physical NIC's (in a dual-NIC system). See: &lt;A href="http://blogs.technet.com/blogfiles/jhoward/WindowsLiveWriter/HowdoesbasicnetworkingworkinHyperV_B633/14_2.jpg" target=_new rel=nofollow&gt;http://blogs.technet.com/blogfiles/jhoward/WindowsLiveWriter/HowdoesbasicnetworkingworkinHyperV_B633/14_2.jpg&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;When I monitor the physical NIC's via performance monitor, they show that they're passing the traffic out of my physical box, to the physical switch (GB full duplex), then back into the box. And BTW - the data is also passing through the virtual switch. Note, this is passing data to/from a WinXP guest. This graphic shows what I'm talking about &lt;A href="http://blogs.technet.com/blogfiles/jhoward/WindowsLiveWriter/HowdoesbasicnetworkingworkinHyperV_B633/13_2.jpg" target=_new rel=nofollow&gt;http://blogs.technet.com/blogfiles/jhoward/WindowsLiveWriter/HowdoesbasicnetworkingworkinHyperV_B633/13_2.jpg&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;#2 - Even though my physical system has dual-GB NIC's, and a full duplex GB switch, the data transfer rate from parent to child partition is equivalent to a poor porforming 100mb network (&amp;lt;5,000,000 bytes/sec). What's up with that? I should be seeing 20+ million bytes/sec on a GB system.&lt;/P&gt;
&lt;P&gt;I've posted this to this blog entry as well as John Howards.&lt;/P&gt;
&lt;P&gt;Any feedback would be great. Network file transfer performance numbers would be wonderful! I'm very excited about Hyper-V, but I'm not sold on performance yet.&lt;/P&gt;
&lt;P&gt;Thanks.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;-----&lt;/P&gt;
&lt;P&gt;[Tony's Reply]&lt;/P&gt;
&lt;P&gt;Performance data is coming...&lt;/P&gt;
&lt;P&gt;&lt;A href="http://blogs.msdn.com/tvoellm/archive/2008/09/24/what-hyper-v-storage-is-best-for-you-show-me-the-numbers.aspx"&gt;http://blogs.msdn.com/tvoellm/archive/2008/09/24/what-hyper-v-storage-is-best-for-you-show-me-the-numbers.aspx&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;As for the networking performance you are seeing from guest to host when using a public switch (is a switch attached to a physical NIC) is you least performant option because you are also depending on your switch to send back the packets.&amp;nbsp; If you want guest to host performance you should use a private switch.&amp;nbsp; There are also factors that you might check out firewall, filtering, ...&amp;nbsp; You also want to make sure you have LSO and TCP offload enabled on the virtual switch.&lt;/P&gt;
&lt;P&gt;You should check out SQL and BizTalk papers...&lt;/P&gt;
&lt;P&gt;&lt;A href="http://blogs.msdn.com/tvoellm/archive/2008/10/07/sql-server-2008-on-windows-server-2008-hyper-v-performance-guidance.aspx"&gt;http://blogs.msdn.com/tvoellm/archive/2008/10/07/sql-server-2008-on-windows-server-2008-hyper-v-performance-guidance.aspx&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;and&lt;/P&gt;
&lt;P&gt;&lt;A href="http://blogs.msdn.com/tvoellm/archive/2008/07/22/biztalk-team-releases-best-practice-doc-for-running-on-ws08-hyper-v.aspx"&gt;http://blogs.msdn.com/tvoellm/archive/2008/07/22/biztalk-team-releases-best-practice-doc-for-running-on-ws08-hyper-v.aspx&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;div style="clear:both;"&gt;&lt;/div&gt;&lt;img src="http://blogs.msdn.com/aggbug.aspx?PostID=8994202" width="1" height="1"&gt;</description></item><item><title>Hyper-V Performance FAQ</title><link>http://blogs.msdn.com/b/tvoellm/archive/2008/04/19/hyper-v-how-to-make-sure-you-are-getting-the-best-performance-when-doing-performance-comparisons.aspx#8576829</link><pubDate>Fri, 06 Jun 2008 06:27:59 GMT</pubDate><guid isPermaLink="false">91d46819-8472-40ad-a661-2c78acb4018c:8576829</guid><dc:creator>ALL TOPICS PERFORMANCE</dc:creator><description>&lt;p&gt;Hyper-V Performance FAQ Anthony F Voellm (aka Tony) 6/5/2008 &lt;a rel="nofollow" target="_new" href="http://blogs.msdn.com/tvoellm"&gt;http://blogs.msdn.com/tvoellm&lt;/a&gt; Q: What is&lt;/p&gt;
&lt;img src="http://blogs.msdn.com/aggbug.aspx?PostID=8576829" width="1" height="1"&gt;</description></item><item><title>re: Hyper-V: How to make sure you are getting the best performance when doing performance comparisons</title><link>http://blogs.msdn.com/b/tvoellm/archive/2008/04/19/hyper-v-how-to-make-sure-you-are-getting-the-best-performance-when-doing-performance-comparisons.aspx#8414588</link><pubDate>Mon, 21 Apr 2008 16:35:41 GMT</pubDate><guid isPermaLink="false">91d46819-8472-40ad-a661-2c78acb4018c:8414588</guid><dc:creator>Pete</dc:creator><description>&lt;P&gt;Graphics for Pitfalls 1 &amp;amp; 2 are switched&lt;/P&gt;
&lt;P&gt;[[Tony- Thanks Pete - Fixed]]&lt;/P&gt;&lt;div style="clear:both;"&gt;&lt;/div&gt;&lt;img src="http://blogs.msdn.com/aggbug.aspx?PostID=8414588" width="1" height="1"&gt;</description></item></channel></rss>