<?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-switching on multi-cpu host in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970566#M120293</link>
    <description>Mikko,&lt;BR /&gt;&lt;BR /&gt;I asked for OS and CPU# because there was is a really old streams pty issue that binds threads accidently to the monarch COU, which is mostly #0 (patched with PHNE_19616 or later). This is obviously not your problem though.&lt;BR /&gt;&lt;BR /&gt;I think you should trust the scheduler. &lt;BR /&gt;&lt;BR /&gt;However, you may try to lock the processes to different CPUs and see if performance increases. I attached a simple program that can be used for this.&lt;BR /&gt;&lt;BR /&gt;Compile it:&lt;BR /&gt;# cc -o mpctl mpctl.c&lt;BR /&gt;&lt;BR /&gt;Bind, e.g. PID 500 to CPU 3:&lt;BR /&gt;# ./mpctl 3 500&lt;BR /&gt;&lt;BR /&gt;Unbind it again:&lt;BR /&gt;# ./mpctl -1 500&lt;BR /&gt;&lt;BR /&gt;Best regards...&lt;BR /&gt; Dietmar.&lt;BR /&gt;&lt;BR /&gt;</description>
    <pubDate>Tue, 13 May 2003 14:55:15 GMT</pubDate>
    <dc:creator>Dietmar Konermann</dc:creator>
    <dc:date>2003-05-13T14:55:15Z</dc:date>
    <item>
      <title>cpu-switching on multi-cpu host</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970559#M120286</link>
      <description>We have four service processes running on four-cpu host (rp-series). 3 of these 4 are running on same cpu all the time. Even if the processor load increases near 100% the processes stay on that same cpu while all the other cpus are in practice idling. How can we balance this load to all processors? Or at least to 2-3 processors?&lt;BR /&gt;&lt;BR /&gt;Is it possible to 'lock' some certain process to some certain processor?</description>
      <pubDate>Mon, 12 May 2003 08:41:20 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970559#M120286</guid>
      <dc:creator>Timo J</dc:creator>
      <dc:date>2003-05-12T08:41:20Z</dc:date>
    </item>
    <item>
      <title>Re: cpu-switching on multi-cpu host</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970560#M120287</link>
      <description>&lt;BR /&gt;check this link&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0xad97663ce855d511abcd0090277a778c,00.html" target="_blank"&gt;http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0xad97663ce855d511abcd0090277a778c,00.html&lt;/A&gt;</description>
      <pubDate>Mon, 12 May 2003 08:49:28 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970560#M120287</guid>
      <dc:creator>T G Manikandan</dc:creator>
      <dc:date>2003-05-12T08:49:28Z</dc:date>
    </item>
    <item>
      <title>Re: cpu-switching on multi-cpu host</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970561#M120288</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;You can use hp PRM (Process Resource Manager)to do this.&lt;BR /&gt;&lt;BR /&gt;but it's surprising to hear that one CPU is used about 100% and other are idle. infact in HP UX kernel will take care of this and i naver saw behaviour like this ...&lt;BR /&gt;&lt;BR /&gt;SUnil</description>
      <pubDate>Mon, 12 May 2003 09:03:27 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970561#M120288</guid>
      <dc:creator>Sunil Sharma_1</dc:creator>
      <dc:date>2003-05-12T09:03:27Z</dc:date>
    </item>
    <item>
      <title>Re: cpu-switching on multi-cpu host</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970562#M120289</link>
      <description>Using PRM one can assign some percentage of CPU to specific processors.It does not lock certain processes to specific CPU.&lt;BR /&gt;&lt;BR /&gt;One can write programs to do that.&lt;BR /&gt;&lt;BR /&gt;check my above link where Bill has an example of that.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Thanks</description>
      <pubDate>Mon, 12 May 2003 09:06:45 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970562#M120289</guid>
      <dc:creator>T G Manikandan</dc:creator>
      <dc:date>2003-05-12T09:06:45Z</dc:date>
    </item>
    <item>
      <title>Re: cpu-switching on multi-cpu host</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970563#M120290</link>
      <description>The HPUX scheduler is designed to do this kind of balancing automatically. PRM is not necessary for this.&lt;BR /&gt;&lt;BR /&gt;Please provide more information....&lt;BR /&gt;&lt;BR /&gt;- What OS revision&lt;BR /&gt;&lt;BR /&gt;- Is always the same CPU loaded? Is it CPU0?&lt;BR /&gt;&lt;BR /&gt;- What means "near 100%"&lt;BR /&gt;  ... maybe balancing is not needed at all?&lt;BR /&gt;&lt;BR /&gt;It is possible (but usually not needed!) to "lock" a process to a CPU using the mpctl() syscall. Beginning with 11.11 you have the mpsched(1) command.&lt;BR /&gt;&lt;BR /&gt;Best regards...&lt;BR /&gt; Dietmar.</description>
      <pubDate>Mon, 12 May 2003 10:37:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970563#M120290</guid>
      <dc:creator>Dietmar Konermann</dc:creator>
      <dc:date>2003-05-12T10:37:05Z</dc:date>
    </item>
    <item>
      <title>Re: cpu-switching on multi-cpu host</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970564#M120291</link>
      <description>HPUX normally manages this if it thinks there is a better assignment for the processes.&lt;BR /&gt;&lt;BR /&gt;So the processes must be running fine on the single processor.&lt;BR /&gt;&lt;BR /&gt;It would be nice to know what OS.&lt;BR /&gt;&lt;BR /&gt;I have seen some anomalies with 11.11 in that a process that takes 100% of a cpu will be shifted between different idle cpus and will not stick to a single cpu.  This causes a slower running process because of the switching.  This did not happen with 10.20</description>
      <pubDate>Mon, 12 May 2003 11:48:07 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970564#M120291</guid>
      <dc:creator>John Bolene</dc:creator>
      <dc:date>2003-05-12T11:48:07Z</dc:date>
    </item>
    <item>
      <title>Re: cpu-switching on multi-cpu host</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970565#M120292</link>
      <description>OS revision is 11.00.&lt;BR /&gt;And the CPU I'm talking is CPU3. (At least, during nighttime when there is very small load per service process, those three processes are running on CPU3). &lt;BR /&gt;&lt;BR /&gt;We started wondering this issue, because one night, one minute after the scheduled backup started, one of our service processes died. At least it _looked_ so obvious that the service process died for some resource problem with the backup.&lt;BR /&gt;&lt;BR /&gt;About the need of the load balancing; I was thinking like Dietmar said but I wasn't just sure; looks like there's no real need to load balancing after all because the load on that one cpu stays between 90-96% during that backup. &lt;BR /&gt;&lt;BR /&gt;It just looks strange that one processor does all the work while the others are idling. &lt;BR /&gt;&lt;BR /&gt;Attachment file contains one snapshot of top output during backup. And if we look the average cpu percentage, everything looks ok. &lt;BR /&gt;&lt;BR /&gt;So, summa summarum; can I trust that the operating system handles perfectly load balancing between cpus when needed without interfering any user processes or do I had to question the functionality of multi-cpu-system ;)&lt;BR /&gt;</description>
      <pubDate>Mon, 12 May 2003 17:12:32 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970565#M120292</guid>
      <dc:creator>Timo J</dc:creator>
      <dc:date>2003-05-12T17:12:32Z</dc:date>
    </item>
    <item>
      <title>Re: cpu-switching on multi-cpu host</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970566#M120293</link>
      <description>Mikko,&lt;BR /&gt;&lt;BR /&gt;I asked for OS and CPU# because there was is a really old streams pty issue that binds threads accidently to the monarch COU, which is mostly #0 (patched with PHNE_19616 or later). This is obviously not your problem though.&lt;BR /&gt;&lt;BR /&gt;I think you should trust the scheduler. &lt;BR /&gt;&lt;BR /&gt;However, you may try to lock the processes to different CPUs and see if performance increases. I attached a simple program that can be used for this.&lt;BR /&gt;&lt;BR /&gt;Compile it:&lt;BR /&gt;# cc -o mpctl mpctl.c&lt;BR /&gt;&lt;BR /&gt;Bind, e.g. PID 500 to CPU 3:&lt;BR /&gt;# ./mpctl 3 500&lt;BR /&gt;&lt;BR /&gt;Unbind it again:&lt;BR /&gt;# ./mpctl -1 500&lt;BR /&gt;&lt;BR /&gt;Best regards...&lt;BR /&gt; Dietmar.&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Tue, 13 May 2003 14:55:15 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cpu-switching-on-multi-cpu-host/m-p/2970566#M120293</guid>
      <dc:creator>Dietmar Konermann</dc:creator>
      <dc:date>2003-05-13T14:55:15Z</dc:date>
    </item>
  </channel>
</rss>

