<?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: Performance PROBLEM on k570 server. in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110706#M148754</link>
    <description>Hi,&lt;BR /&gt;&lt;BR /&gt;first quick guess would be that you desperately need more memory. However, you need to invesigate a bit deeper see attached doc.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;Bernhard</description>
    <pubDate>Wed, 05 Nov 2003 05:58:06 GMT</pubDate>
    <dc:creator>Bernhard Mueller</dc:creator>
    <dc:date>2003-11-05T05:58:06Z</dc:date>
    <item>
      <title>Performance PROBLEM on k570 server.</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110704#M148752</link>
      <description>Hi Gurus.&lt;BR /&gt;I need your help....&lt;BR /&gt;I have a K570 server with 5 X 200MHz processors, 2 gig memory, 2 storage arrays of + - 200gig in both, which we use for our in-house application.&lt;BR /&gt; &lt;BR /&gt;I also have another serverwith the same config, but just an extra cpu, which we use for our database(sybase) server.&lt;BR /&gt;&lt;BR /&gt;Now we have some major performance issues on our application server for the past 2 weeks. Nothing new was added. The performance has gradually been degrading as we keep on adding more users to the server. See attached document for an output of top.&lt;BR /&gt;&lt;BR /&gt;I think our server is over utilised, but not sure as the idle time is sometims ok.&lt;BR /&gt;&lt;BR /&gt;If there are any advise, please do not hesitate..&lt;BR /&gt;Thanks in advance.&lt;BR /&gt;</description>
      <pubDate>Wed, 05 Nov 2003 05:16:55 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110704#M148752</guid>
      <dc:creator>Shaamil</dc:creator>
      <dc:date>2003-11-05T05:16:55Z</dc:date>
    </item>
    <item>
      <title>Re: Performance PROBLEM on k570 server.</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110705#M148753</link>
      <description>Do you have a baseline (top, etc ...) for this server when everything was fine ?&lt;BR /&gt;&lt;BR /&gt;swapinfo -tam output.&lt;BR /&gt;Is this server application or application + database ?&lt;BR /&gt;&lt;BR /&gt;more info available ?&lt;BR /&gt;&lt;BR /&gt;Rgds,&lt;BR /&gt;JL&lt;BR /&gt;</description>
      <pubDate>Wed, 05 Nov 2003 05:39:17 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110705#M148753</guid>
      <dc:creator>Jean-Luc Oudart</dc:creator>
      <dc:date>2003-11-05T05:39:17Z</dc:date>
    </item>
    <item>
      <title>Re: Performance PROBLEM on k570 server.</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110706#M148754</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;first quick guess would be that you desperately need more memory. However, you need to invesigate a bit deeper see attached doc.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;Bernhard</description>
      <pubDate>Wed, 05 Nov 2003 05:58:06 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110706#M148754</guid>
      <dc:creator>Bernhard Mueller</dc:creator>
      <dc:date>2003-11-05T05:58:06Z</dc:date>
    </item>
    <item>
      <title>Re: Performance PROBLEM on k570 server.</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110707#M148755</link>
      <description>Your load average is huge.&lt;BR /&gt;&lt;BR /&gt;This means that your CPUs can't keep up with all the processes (3700) that are ready to execute.&lt;BR /&gt;&lt;BR /&gt;I also note that swapper is running and your free memory is low.  This probably means that your buffer cache is fully decreased.  I guess you also see vhand running a lot.&lt;BR /&gt;&lt;BR /&gt;ttisr services client requests.  This is another symptom of too many interactive users all competing for resources.&lt;BR /&gt;&lt;BR /&gt;You need to add memory and upgrade to 4 of the fastest cpus your machine can hold, otherwise upgade the whole machine.&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Wed, 05 Nov 2003 06:43:57 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110707#M148755</guid>
      <dc:creator>Steve Lewis</dc:creator>
      <dc:date>2003-11-05T06:43:57Z</dc:date>
    </item>
    <item>
      <title>Re: Performance PROBLEM on k570 server.</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110708#M148756</link>
      <description>I thank you.  I used to think I had it bad here.&lt;BR /&gt;&lt;BR /&gt;We run multiple instances of Oracle off of K370's - about a dozen apiece.  The users complain of "almost unusable" slowness when load hits 4 or 5, never mind 50.&lt;BR /&gt;&lt;BR /&gt;It's the users.  You are CPU bound, and your poor old K cannot keep up.  Adding the extra processor will have some effect, but certainly will not solve the problem. &lt;BR /&gt;&lt;BR /&gt;If this load is constant, you need a more powerful box.  There is a chance that you have something in particular going on that is tying up the CPU's - in my attachment, it's Omniback, but during our afternoon peak usage, things look the same.</description>
      <pubDate>Wed, 05 Nov 2003 07:51:03 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-on-k570-server/m-p/3110708#M148756</guid>
      <dc:creator>Jon Mattatall</dc:creator>
      <dc:date>2003-11-05T07:51:03Z</dc:date>
    </item>
  </channel>
</rss>

