<?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: vmstat and acceptable value for cs (context switches) in Operating System - Linux</title>
    <link>https://community.hpe.com/t5/operating-system-linux/vmstat-and-acceptable-value-for-cs-context-switches/m-p/4197214#M62087</link>
    <description>Hmmm, this sounds like a kind of "tell me your need, i'll show you how to do without" answer.&lt;BR /&gt;&lt;BR /&gt;I indeed also use other metrics.&lt;BR /&gt;&lt;BR /&gt;But the values of the run queue AND cswitch/s can tell me if it is better to add more processors or to put a more powerfull processor.&lt;BR /&gt;&lt;BR /&gt;Thanks, anyway.&lt;BR /&gt;&lt;BR /&gt;Louis Coilliot</description>
    <pubDate>Tue, 13 May 2008 21:35:04 GMT</pubDate>
    <dc:creator>Louis Coilliot</dc:creator>
    <dc:date>2008-05-13T21:35:04Z</dc:date>
    <item>
      <title>vmstat and acceptable value for cs (context switches)</title>
      <link>https://community.hpe.com/t5/operating-system-linux/vmstat-and-acceptable-value-for-cs-context-switches/m-p/4197212#M62085</link>
      <description>Hello,&lt;BR /&gt;&lt;BR /&gt;I wonder what is the range for acceptable values of context switches for up-to date Intel x86 processors (let's say, a Xeon® X5460 Quad Core 3.16 Ghz for example).&lt;BR /&gt;&lt;BR /&gt;I can't find any recent information about this.&lt;BR /&gt;&lt;BR /&gt;The fresher information I found is an article of 2002 that says (in substance):&lt;BR /&gt;A lot of context switching = greater than 20 000 per second.&lt;BR /&gt;&lt;BR /&gt;If I apply Moore's Law, that would make 160 000&lt;BR /&gt;for a CPU in 2008.&lt;BR /&gt;&lt;BR /&gt;Does anyone has informations on this ?&lt;BR /&gt;&lt;BR /&gt;Let me know, thanks.&lt;BR /&gt;&lt;BR /&gt;Louis Coilliot&lt;BR /&gt;</description>
      <pubDate>Tue, 13 May 2008 20:45:28 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/vmstat-and-acceptable-value-for-cs-context-switches/m-p/4197212#M62085</guid>
      <dc:creator>Louis Coilliot</dc:creator>
      <dc:date>2008-05-13T20:45:28Z</dc:date>
    </item>
    <item>
      <title>Re: vmstat and acceptable value for cs (context switches)</title>
      <link>https://community.hpe.com/t5/operating-system-linux/vmstat-and-acceptable-value-for-cs-context-switches/m-p/4197213#M62086</link>
      <description>The true is that, I never used CS as a performance parameter. I don't think that there is a "suggested" value, it depends of the system, drivers and applications.&lt;BR /&gt;&lt;BR /&gt;Focus on other performance counters.</description>
      <pubDate>Tue, 13 May 2008 21:24:58 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/vmstat-and-acceptable-value-for-cs-context-switches/m-p/4197213#M62086</guid>
      <dc:creator>Ivan Ferreira</dc:creator>
      <dc:date>2008-05-13T21:24:58Z</dc:date>
    </item>
    <item>
      <title>Re: vmstat and acceptable value for cs (context switches)</title>
      <link>https://community.hpe.com/t5/operating-system-linux/vmstat-and-acceptable-value-for-cs-context-switches/m-p/4197214#M62087</link>
      <description>Hmmm, this sounds like a kind of "tell me your need, i'll show you how to do without" answer.&lt;BR /&gt;&lt;BR /&gt;I indeed also use other metrics.&lt;BR /&gt;&lt;BR /&gt;But the values of the run queue AND cswitch/s can tell me if it is better to add more processors or to put a more powerfull processor.&lt;BR /&gt;&lt;BR /&gt;Thanks, anyway.&lt;BR /&gt;&lt;BR /&gt;Louis Coilliot</description>
      <pubDate>Tue, 13 May 2008 21:35:04 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/vmstat-and-acceptable-value-for-cs-context-switches/m-p/4197214#M62087</guid>
      <dc:creator>Louis Coilliot</dc:creator>
      <dc:date>2008-05-13T21:35:04Z</dc:date>
    </item>
    <item>
      <title>Re: vmstat and acceptable value for cs (context switches)</title>
      <link>https://community.hpe.com/t5/operating-system-linux/vmstat-and-acceptable-value-for-cs-context-switches/m-p/4197215#M62088</link>
      <description>I agree tha runq is a metric, related with idle CPU and load average. But still I don't know how CS can give you good/bad values.</description>
      <pubDate>Wed, 14 May 2008 13:34:37 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/vmstat-and-acceptable-value-for-cs-context-switches/m-p/4197215#M62088</guid>
      <dc:creator>Ivan Ferreira</dc:creator>
      <dc:date>2008-05-14T13:34:37Z</dc:date>
    </item>
  </channel>
</rss>

