<?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: Load average recommended thresholds in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261398#M613361</link>
    <description>The CPU load is a very misleading metric including its name.&lt;BR /&gt;&lt;BR /&gt;You can have a load below 1 and have a serious bottleneck and vice-versa you can have a load of 20 and have a very responsive system.&lt;BR /&gt;&lt;BR /&gt;Look at Bill's responses in these threads&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=750840" target="_blank"&gt;http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=750840&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=876751" target="_blank"&gt;http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=876751&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=452648" target="_blank"&gt;http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=452648&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1071636" target="_blank"&gt;http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1071636&lt;/A&gt;</description>
    <pubDate>Wed, 03 Nov 2010 12:24:53 GMT</pubDate>
    <dc:creator>TTr</dc:creator>
    <dc:date>2010-11-03T12:24:53Z</dc:date>
    <item>
      <title>Load average recommended thresholds</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261396#M613359</link>
      <description>Hi all,&lt;BR /&gt;&lt;BR /&gt;I need to establish a policy for monitoring my HP-UX systems and I’m trying to figure out which would be the recommended thresholds for warning and critical load average alerts.&lt;BR /&gt;&lt;BR /&gt;After having read a lot of documentation it’s not very clear for me what would be recommended values from top’s load averages.&lt;BR /&gt;&lt;BR /&gt;Some people recommend maximum load averages of 3 or 4 and some other say there is a problem on the system only from values of 14-15 on.&lt;BR /&gt;&lt;BR /&gt;I have, for example, a HP-UX 11.31 running on a ia64 hp server BL870c with following characteristics:&lt;BR /&gt;&lt;BR /&gt;CPU info:&lt;BR /&gt;  4 Intel(R) Itanium 2 9100 series processors (1.59 GHz, 18 MB)&lt;BR /&gt;          532 MT/s bus, CPU version A1&lt;BR /&gt;          16 logical processors (4 per socket)&lt;BR /&gt;&lt;BR /&gt;Memory: 131041 MB (127.97 GB)&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Top’s output usually shows values like following ones:&lt;BR /&gt;&lt;BR /&gt;Load averages: 0.35, 0.39, 0.42&lt;BR /&gt;826 processes: 737 sleeping, 87 running, 2 zombies&lt;BR /&gt;Cpu states:&lt;BR /&gt;CPU   LOAD   USER   NICE    SYS   IDLE  BLOCK  SWAIT   INTR   SSYS&lt;BR /&gt; 0    0.46  29.7%   0.0%   6.9%  63.4%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt; 1    0.25  41.6%   0.0%   8.9%  49.5%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt; 2    0.47  23.8%   0.0%   5.0%  71.3%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt; 3    0.26  47.5%   0.0%   6.9%  45.5%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt; 4    0.50  25.7%   0.0%   9.9%  64.4%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt; 5    0.24  21.8%   0.0%   3.0%  75.2%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt; 6    0.42  15.8%   0.0%   7.9%  76.2%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt; 7    0.26  15.0%   0.0%   6.0%  79.0%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt; 8    0.44  29.0%   0.0%   6.0%  65.0%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt; 9    0.27  23.0%   0.0%   2.0%  75.0%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt;10    0.42  74.0%   0.0%   2.0%  24.0%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt;11    0.25  21.0%   0.0%   3.0%  76.0%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt;12    0.40  20.0%   0.0%   7.0%  73.0%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt;13    0.26  17.0%   0.0%   5.0%  78.0%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt;14    0.42  97.0%   0.0%   0.0%   3.0%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt;15    0.26   5.0%   0.0%   1.0%  94.0%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt;---   ----  -----  -----  -----  -----  -----  -----  -----  -----&lt;BR /&gt;avg   0.35  31.7%   0.0%   5.0%  63.4%   0.0%   0.0%   0.0%   0.0%&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;I’d thank if somebody could put some light on all this and let me know when could I run into trouble with load average. Of course, I should consider some other facts like slowness of system, etc, etc, but for the moment I need to take a specific value to set my alerts. &lt;BR /&gt;&lt;BR /&gt;Thanks in advance.&lt;BR /&gt;</description>
      <pubDate>Wed, 03 Nov 2010 11:21:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261396#M613359</guid>
      <dc:creator>luissma</dc:creator>
      <dc:date>2010-11-03T11:21:05Z</dc:date>
    </item>
    <item>
      <title>Re: Load average recommended thresholds</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261397#M613360</link>
      <description>Your current status ,load average is good (very low)&lt;BR /&gt;&lt;BR /&gt;If you use (if you have) glance tool ,you can use its default, to analyze load average.&lt;BR /&gt;By default cpu bottleneck is around 50% ,alarm will be triggered as warning in glance.&lt;BR /&gt;&lt;BR /&gt;additionaly,   On multi-processor systems, the values represent the average over all CPUs. &lt;BR /&gt;Thus the percentage  columns never exceed 100. For individual processor detail, &lt;BR /&gt;use the 'a' (CPU By Processor) screen. &lt;BR /&gt;</description>
      <pubDate>Wed, 03 Nov 2010 11:41:18 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261397#M613360</guid>
      <dc:creator>Hakki Aydin Ucar</dc:creator>
      <dc:date>2010-11-03T11:41:18Z</dc:date>
    </item>
    <item>
      <title>Re: Load average recommended thresholds</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261398#M613361</link>
      <description>The CPU load is a very misleading metric including its name.&lt;BR /&gt;&lt;BR /&gt;You can have a load below 1 and have a serious bottleneck and vice-versa you can have a load of 20 and have a very responsive system.&lt;BR /&gt;&lt;BR /&gt;Look at Bill's responses in these threads&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=750840" target="_blank"&gt;http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=750840&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=876751" target="_blank"&gt;http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=876751&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=452648" target="_blank"&gt;http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=452648&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1071636" target="_blank"&gt;http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1071636&lt;/A&gt;</description>
      <pubDate>Wed, 03 Nov 2010 12:24:53 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261398#M613361</guid>
      <dc:creator>TTr</dc:creator>
      <dc:date>2010-11-03T12:24:53Z</dc:date>
    </item>
    <item>
      <title>Re: Load average recommended thresholds</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261399#M613362</link>
      <description>I have always hated the load average output. It may be because I have known old school admins that place to much emphasis on it. From my understanding you are possibly hitting a bottleneck if your load average is nearing the same amount you have in cpus/cores. So if your load average starts getting close to 16, then you might have a problem. I say might because it is not necessarily bad that your processors are doing work. Tuning is a science and an art. You really how the system is supposed to function and what your users are expecting. I would suggest getting some baseline data and making comparisons along the way. &lt;BR /&gt;&lt;BR /&gt;It is hard to give an exact number since every server/application is a little different. There are no hard set numbers to give anyone. That is why performance/tuning guys make lots o' cash. In the words of any good tuning guru, the answer is "It depends".</description>
      <pubDate>Wed, 03 Nov 2010 19:46:57 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261399#M613362</guid>
      <dc:creator>Court Campbell</dc:creator>
      <dc:date>2010-11-03T19:46:57Z</dc:date>
    </item>
    <item>
      <title>Re: Load average recommended thresholds</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261400#M613363</link>
      <description>Understood. Thanks to all, including Bill Hassell :)&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;&lt;BR /&gt;Luisma Arranz</description>
      <pubDate>Thu, 04 Nov 2010 08:52:41 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/load-average-recommended-thresholds/m-p/5261400#M613363</guid>
      <dc:creator>luissma</dc:creator>
      <dc:date>2010-11-04T08:52:41Z</dc:date>
    </item>
  </channel>
</rss>

