<?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: cpu load in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649493#M803397</link>
    <description>Quite often, a high load means that a series of short-lived processes (perhaps a set of polling programs) are running. They are assigned the first processor, run for a few milliseconds then go to sleep. You can see this as the %system overhead will be higher than the typical 3-8%. Note that in a multi-processor system, one process cannot be shared across multiple CPUs. A threaded process (requires special design) can indeed be spread over several processors.</description>
    <pubDate>Fri, 14 Oct 2005 07:50:45 GMT</pubDate>
    <dc:creator>Bill Hassell</dc:creator>
    <dc:date>2005-10-14T07:50:45Z</dc:date>
    <item>
      <title>cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649485#M803389</link>
      <description>I have a L1500 with 2 processors.&lt;BR /&gt;One of the processors always has a load of 4, and the other one 0.15.&lt;BR /&gt;How can that be? And how do I change it?&lt;BR /&gt;&lt;BR /&gt;Greetings, Alfons Velthof</description>
      <pubDate>Fri, 14 Oct 2005 04:58:43 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649485#M803389</guid>
      <dc:creator>A.G.M. Velthof</dc:creator>
      <dc:date>2005-10-14T04:58:43Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649486#M803390</link>
      <description>&lt;A href="http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=857310" target="_blank"&gt;http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=857310&lt;/A&gt;&lt;BR /&gt;&lt;A href="http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=723070" target="_blank"&gt;http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=723070&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;Check these threads, it could be helpful&lt;BR /&gt;&lt;BR /&gt;-Arun</description>
      <pubDate>Fri, 14 Oct 2005 05:11:20 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649486#M803390</guid>
      <dc:creator>Arunvijai_4</dc:creator>
      <dc:date>2005-10-14T05:11:20Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649487#M803391</link>
      <description>It's probably due to whatever your application is being single threaded - i.e. it cannot take advantage of multiple CPUs and is therefore running on just one. What apps are running on your server?&lt;BR /&gt;&lt;BR /&gt;Regards - Lee</description>
      <pubDate>Fri, 14 Oct 2005 05:11:40 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649487#M803391</guid>
      <dc:creator>Lee Harris_5</dc:creator>
      <dc:date>2005-10-14T05:11:40Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649488#M803392</link>
      <description>Load average in itself os not an indication of how system is performing. Rather, priority job queue is important. Look at it-glance/gpm.&lt;BR /&gt;&lt;BR /&gt;Also as said, single threaded processes/apps/programs will bind to a one cpu and cause this.</description>
      <pubDate>Fri, 14 Oct 2005 05:21:10 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649488#M803392</guid>
      <dc:creator>RAC_1</dc:creator>
      <dc:date>2005-10-14T05:21:10Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649489#M803393</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;here is the same prob. May be help ful to you.Just go through..&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=570911" target="_blank"&gt;http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=570911&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;Syam</description>
      <pubDate>Fri, 14 Oct 2005 05:28:19 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649489#M803393</guid>
      <dc:creator>Ranjith_5</dc:creator>
      <dc:date>2005-10-14T05:28:19Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649490#M803394</link>
      <description>Oracle is running on the system.&lt;BR /&gt;And Oracle Interconnect.</description>
      <pubDate>Fri, 14 Oct 2005 06:11:26 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649490#M803394</guid>
      <dc:creator>A.G.M. Velthof</dc:creator>
      <dc:date>2005-10-14T06:11:26Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649491#M803395</link>
      <description>You seem to have an application which is using the CPU rather heavily. &lt;BR /&gt;&lt;BR /&gt;If the application is not threaded, it can utilize only one CPU at a time. Effectively, one processor is busy running the application, so all the OS tasks run on the other processor. Many old applications are like this.&lt;BR /&gt;&lt;BR /&gt;I'm afraid the only solution would be to re-implement the application to use several threads. &lt;BR /&gt;</description>
      <pubDate>Fri, 14 Oct 2005 06:12:43 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649491#M803395</guid>
      <dc:creator>Matti_Kurkela</dc:creator>
      <dc:date>2005-10-14T06:12:43Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649492#M803396</link>
      <description>Perhaps, you have to use Glance plus instead of top to dig more detail into the problem. &lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://h21007.www2.hp.com/dspp/dld/dld_DownloadsListingPage_IDX/1,2381,11169,00.html" target="_blank"&gt;http://h21007.www2.hp.com/dspp/dld/dld_DownloadsListingPage_IDX/1,2381,11169,00.html&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;-Arun</description>
      <pubDate>Fri, 14 Oct 2005 06:58:45 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649492#M803396</guid>
      <dc:creator>Arunvijai_4</dc:creator>
      <dc:date>2005-10-14T06:58:45Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649493#M803397</link>
      <description>Quite often, a high load means that a series of short-lived processes (perhaps a set of polling programs) are running. They are assigned the first processor, run for a few milliseconds then go to sleep. You can see this as the %system overhead will be higher than the typical 3-8%. Note that in a multi-processor system, one process cannot be shared across multiple CPUs. A threaded process (requires special design) can indeed be spread over several processors.</description>
      <pubDate>Fri, 14 Oct 2005 07:50:45 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649493#M803397</guid>
      <dc:creator>Bill Hassell</dc:creator>
      <dc:date>2005-10-14T07:50:45Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649494#M803398</link>
      <description>Because of the interconnect adapters, about 10 "java" processes are running and consuming most of the cpu resources, but according to "top" all are running on the processor 0.&lt;BR /&gt;</description>
      <pubDate>Fri, 14 Oct 2005 08:27:00 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649494#M803398</guid>
      <dc:creator>A.G.M. Velthof</dc:creator>
      <dc:date>2005-10-14T08:27:00Z</dc:date>
    </item>
    <item>
      <title>Re: cpu load</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649495#M803399</link>
      <description>That sounds quite normal. The reason they seem to hogging the CPU is that the Java threads are very short lived. top (and Glance and other tools) can't represent dozens of short lived processes in a meaningful way. During the measurement period, 1 second for instance, 10 processes are shown aas running on CPU 0, but really, each process ran for a very short time, then probably went into a wait state for a semaphore or poll signal.</description>
      <pubDate>Fri, 14 Oct 2005 13:09:53 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-load/m-p/3649495#M803399</guid>
      <dc:creator>Bill Hassell</dc:creator>
      <dc:date>2005-10-14T13:09:53Z</dc:date>
    </item>
  </channel>
</rss>

