<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: Proliant BL20p G2 and high CPU load/queue length in ProLiant Servers (ML,DL,SL)</title>
    <link>https://community.hpe.com/t5/proliant-servers-ml-dl-sl/proliant-bl20p-g2-and-high-cpu-load-queue-length/m-p/4044434#M67624</link>
    <description>Disabling HyperThreading did not help. Second thing i tried is i used Interrupt-Affinity Filter Tool (&lt;A href="http://support.microsoft.com/kb/252867)" target="_blank"&gt;http://support.microsoft.com/kb/252867)&lt;/A&gt; to set that Nic1 uses CPU2 and Nic2 users CPU3, now the interrupts/sec are balanced a bit more evenly over cpu's. &lt;BR /&gt;&lt;BR /&gt;Client's are using this terminal to browse internet with Internet Explorer 6, so i guess there is big network load.</description>
    <pubDate>Fri, 27 Jul 2007 03:21:17 GMT</pubDate>
    <dc:creator>Kalle Pihelgas</dc:creator>
    <dc:date>2007-07-27T03:21:17Z</dc:date>
    <item>
      <title>Proliant BL20p G2 and high CPU load/queue length</title>
      <link>https://community.hpe.com/t5/proliant-servers-ml-dl-sl/proliant-bl20p-g2-and-high-cpu-load-queue-length/m-p/4044432#M67622</link>
      <description>hi, &lt;BR /&gt;&lt;BR /&gt;I have problem with high CPU load and queue length with Proliant BL20p G2 (2x3GHZ CPU HT, 8GB memory, 2 nic-s in teaming, MS Windows Server 2003 Ent SP2). Server is used as MS terminal server. &lt;BR /&gt;&lt;BR /&gt;Problem seems to be that most interrupts are using CPU0 instead of spreading to CPU0,CPU1,CPU2,CPU3. &lt;BR /&gt;&lt;BR /&gt;Data collected with performance monitor shows following: &lt;BR /&gt;&lt;BR /&gt;CPU0 load: 100&lt;BR /&gt;CPU0 interrupts/sec: over 2000&lt;BR /&gt;CPU1 load: 100&lt;BR /&gt;CPU1 interrupts/sec: 64&lt;BR /&gt;CPU2 load: 100&lt;BR /&gt;CPU2 interrupts/sec: 64&lt;BR /&gt;CPU3 load: 100&lt;BR /&gt;CPU3 interrupts/sec: 64&lt;BR /&gt;&lt;BR /&gt;CPU Queue length : 74&lt;BR /&gt;Total number of active terminal sessions: 60&lt;BR /&gt;&lt;BR /&gt;System firmware's are all up-to date. Hp Support Pack is 7.80. &lt;BR /&gt;&lt;BR /&gt;Any help and hints what to do next are welcome.</description>
      <pubDate>Thu, 26 Jul 2007 06:08:17 GMT</pubDate>
      <guid>https://community.hpe.com/t5/proliant-servers-ml-dl-sl/proliant-bl20p-g2-and-high-cpu-load-queue-length/m-p/4044432#M67622</guid>
      <dc:creator>Kalle Pihelgas</dc:creator>
      <dc:date>2007-07-26T06:08:17Z</dc:date>
    </item>
    <item>
      <title>Re: Proliant BL20p G2 and high CPU load/queue length</title>
      <link>https://community.hpe.com/t5/proliant-servers-ml-dl-sl/proliant-bl20p-g2-and-high-cpu-load-queue-length/m-p/4044433#M67623</link>
      <description>My first instinct is to disable HyperThreading, but you can get some better info on configuring and sizing by visiting &lt;A href="http://www.hp.com/servers/activeanswers" target="_blank"&gt;http://www.hp.com/servers/activeanswers&lt;/A&gt; --&amp;gt; Business Solutions --&amp;gt; Server Based Computing and then looking for documents pertaining to Terminal Services including a sizing utility.</description>
      <pubDate>Thu, 26 Jul 2007 10:27:31 GMT</pubDate>
      <guid>https://community.hpe.com/t5/proliant-servers-ml-dl-sl/proliant-bl20p-g2-and-high-cpu-load-queue-length/m-p/4044433#M67623</guid>
      <dc:creator>David Claypool</dc:creator>
      <dc:date>2007-07-26T10:27:31Z</dc:date>
    </item>
    <item>
      <title>Re: Proliant BL20p G2 and high CPU load/queue length</title>
      <link>https://community.hpe.com/t5/proliant-servers-ml-dl-sl/proliant-bl20p-g2-and-high-cpu-load-queue-length/m-p/4044434#M67624</link>
      <description>Disabling HyperThreading did not help. Second thing i tried is i used Interrupt-Affinity Filter Tool (&lt;A href="http://support.microsoft.com/kb/252867)" target="_blank"&gt;http://support.microsoft.com/kb/252867)&lt;/A&gt; to set that Nic1 uses CPU2 and Nic2 users CPU3, now the interrupts/sec are balanced a bit more evenly over cpu's. &lt;BR /&gt;&lt;BR /&gt;Client's are using this terminal to browse internet with Internet Explorer 6, so i guess there is big network load.</description>
      <pubDate>Fri, 27 Jul 2007 03:21:17 GMT</pubDate>
      <guid>https://community.hpe.com/t5/proliant-servers-ml-dl-sl/proliant-bl20p-g2-and-high-cpu-load-queue-length/m-p/4044434#M67624</guid>
      <dc:creator>Kalle Pihelgas</dc:creator>
      <dc:date>2007-07-27T03:21:17Z</dc:date>
    </item>
  </channel>
</rss>

