<?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 usage for a specific process in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415514#M861322</link>
    <description>Hi Sri/Patrick,&lt;BR /&gt;Thanks for the answer.The command works fine.&lt;BR /&gt;I've already configured the gpm and 'm using perfiew to monitor the usage.I wanted my own script to monitor this, and this works fine.&lt;BR /&gt;Now I've one more question.When i monitor apache using apache status tool, the cpu load is shown as 80% and the calculated %cpu value using the ps command is 0.21%.&lt;BR /&gt;The cpu load sometimes goin to 120% also.So how to interpret the cpu load from apache and %cpu calculated using the command given ?&lt;BR /&gt;Thanks,&lt;BR /&gt;Bijoy&lt;BR /&gt;</description>
    <pubDate>Thu, 04 Nov 2004 23:26:46 GMT</pubDate>
    <dc:creator>Bijoy_1</dc:creator>
    <dc:date>2004-11-04T23:26:46Z</dc:date>
    <item>
      <title>%CPU usage for a specific process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415511#M861319</link>
      <description>Hello friends,&lt;BR /&gt;How to calculate the %cpu usage for a particular process.I'm running apache.There are around 16 httpd process running at a given time.I want to find the aggregate of %cpu fot this.Top is not listing the %cpu usage for this process, and the utilisation is very less.Any help will be appreciated.&lt;BR /&gt;Thanks,</description>
      <pubDate>Thu, 04 Nov 2004 22:20:27 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415511#M861319</guid>
      <dc:creator>Bijoy_1</dc:creator>
      <dc:date>2004-11-04T22:20:27Z</dc:date>
    </item>
    <item>
      <title>Re: %CPU usage for a specific process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415512#M861320</link>
      <description>I would inverst in the GlancePlus Pak software for this.  With the Glance software you can see the amount of CPU Usage for all processes at that time.  The Measureware software will keep a history of usage so you can go back a few days, weeks, or months (depending on how big you make your log files) and see historical process usage.&lt;BR /&gt;&lt;BR /&gt;You can find more information on GlancePlus Pak here:&lt;BR /&gt;&lt;A href="http://www.openview.hp.com/products/gppak2k/index.html" target="_blank"&gt;http://www.openview.hp.com/products/gppak2k/index.html&lt;/A&gt;</description>
      <pubDate>Thu, 04 Nov 2004 22:27:16 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415512#M861320</guid>
      <dc:creator>Patrick Wallek</dc:creator>
      <dc:date>2004-11-04T22:27:16Z</dc:date>
    </item>
    <item>
      <title>Re: %CPU usage for a specific process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415513#M861321</link>
      <description>Hi Bijoy,&lt;BR /&gt;&lt;BR /&gt;The best way is to use 'measureware'. If you have it, group all 'httpd' processes together as an 'application' in /var/opt/perf/parm file and restart measureware. Once the datacollection is started, you can extract the stats using the command 'extract'. It's a easy-to-use tool. There are some examples in parm file so it shouldn't be a problem for you to configure an additional application.&lt;BR /&gt;&lt;BR /&gt;If you don't have measureware, you can use XPG's version of 'ps' to collect the samples.  &lt;BR /&gt;&lt;BR /&gt;UNIX95= ps -C "httpd" -o "pcpu" |sed '1d' |awk '{sum += $1} END {print sum}'&lt;BR /&gt;&lt;BR /&gt;However, customizing measureware is the best solution as you can get quite few metrics in addition to CPU usage.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;-Sri</description>
      <pubDate>Thu, 04 Nov 2004 22:32:46 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415513#M861321</guid>
      <dc:creator>Sridhar Bhaskarla</dc:creator>
      <dc:date>2004-11-04T22:32:46Z</dc:date>
    </item>
    <item>
      <title>Re: %CPU usage for a specific process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415514#M861322</link>
      <description>Hi Sri/Patrick,&lt;BR /&gt;Thanks for the answer.The command works fine.&lt;BR /&gt;I've already configured the gpm and 'm using perfiew to monitor the usage.I wanted my own script to monitor this, and this works fine.&lt;BR /&gt;Now I've one more question.When i monitor apache using apache status tool, the cpu load is shown as 80% and the calculated %cpu value using the ps command is 0.21%.&lt;BR /&gt;The cpu load sometimes goin to 120% also.So how to interpret the cpu load from apache and %cpu calculated using the command given ?&lt;BR /&gt;Thanks,&lt;BR /&gt;Bijoy&lt;BR /&gt;</description>
      <pubDate>Thu, 04 Nov 2004 23:26:46 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415514#M861322</guid>
      <dc:creator>Bijoy_1</dc:creator>
      <dc:date>2004-11-04T23:26:46Z</dc:date>
    </item>
    <item>
      <title>Re: %CPU usage for a specific process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415515#M861323</link>
      <description>system load and process working pcu percentage are different :&lt;BR /&gt;. pcpu is the time the cpu have been used for the process for the time it has been given the cpu. Due to multi-thread processes, and to the fact, that a line in top is a process (maybe multi-thread), a process can be shown using 120% (one thread using 80% other 40%).&lt;BR /&gt;. System load is the number of processes that have been in a running state during the mesure time. 5 processes having a pcpu of 100% during the mesure time will give a load of 500% (5).&lt;BR /&gt;&lt;BR /&gt;Note that for a multi-cpu box, you will have a difference between load and load average, where load average is the average of the loads of each processor.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;&lt;BR /&gt;Fred&lt;BR /&gt;</description>
      <pubDate>Fri, 05 Nov 2004 01:44:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-usage-for-a-specific-process/m-p/3415515#M861323</guid>
      <dc:creator>Fred Ruffet</dc:creator>
      <dc:date>2004-11-05T01:44:05Z</dc:date>
    </item>
  </channel>
</rss>

