<?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 Benchmark tests in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450512#M10808</link>
    <description>We are running performance benchmarks on HP-UX 11.0 with Oracle 8.1.5 and MQSeries 5.1.  Certain tests are only to MQSeries and others only to Oracle.&lt;BR /&gt;&lt;BR /&gt;There is one strange consistency between these tests.  Every so many transactions takes 8 seconds whereas the rest takes 0.05 to 0.4 seconds per transaction.&lt;BR /&gt;&lt;BR /&gt;Any ideas where I can start looking.  These transactions which take 8 seconds are not evenly spaced and sometimes in a batch of 100 transactions none occur and yet other batches up to 6 take 8 seconds.&lt;BR /&gt;&lt;BR /&gt;Thank you.</description>
    <pubDate>Thu, 05 Oct 2000 10:29:51 GMT</pubDate>
    <dc:creator>Carlo Henrico_1</dc:creator>
    <dc:date>2000-10-05T10:29:51Z</dc:date>
    <item>
      <title>Benchmark tests</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450512#M10808</link>
      <description>We are running performance benchmarks on HP-UX 11.0 with Oracle 8.1.5 and MQSeries 5.1.  Certain tests are only to MQSeries and others only to Oracle.&lt;BR /&gt;&lt;BR /&gt;There is one strange consistency between these tests.  Every so many transactions takes 8 seconds whereas the rest takes 0.05 to 0.4 seconds per transaction.&lt;BR /&gt;&lt;BR /&gt;Any ideas where I can start looking.  These transactions which take 8 seconds are not evenly spaced and sometimes in a batch of 100 transactions none occur and yet other batches up to 6 take 8 seconds.&lt;BR /&gt;&lt;BR /&gt;Thank you.</description>
      <pubDate>Thu, 05 Oct 2000 10:29:51 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450512#M10808</guid>
      <dc:creator>Carlo Henrico_1</dc:creator>
      <dc:date>2000-10-05T10:29:51Z</dc:date>
    </item>
    <item>
      <title>Re: Benchmark tests</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450513#M10809</link>
      <description>Keeping a close eye on system performance with 'glance' would be my first suggestion. &lt;BR /&gt;&lt;BR /&gt;If you haven't purchased it, you can install a trial version from the Applications CD disk 2.</description>
      <pubDate>Thu, 05 Oct 2000 10:39:17 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450513#M10809</guid>
      <dc:creator>John Palmer</dc:creator>
      <dc:date>2000-10-05T10:39:17Z</dc:date>
    </item>
    <item>
      <title>Re: Benchmark tests</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450514#M10810</link>
      <description>Thanks John, however, my Glance is open permanently with between 1 and 5 second interval time.  According to Glance the system seems to be cruising along at a steady 95%+ idle.&lt;BR /&gt;&lt;BR /&gt;We do have an HPARRAY 12H - I have a suspicion there may be something funny there.</description>
      <pubDate>Thu, 05 Oct 2000 11:01:10 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450514#M10810</guid>
      <dc:creator>Carlo Henrico_1</dc:creator>
      <dc:date>2000-10-05T11:01:10Z</dc:date>
    </item>
    <item>
      <title>Re: Benchmark tests</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450515#M10811</link>
      <description>Well it certainly doesn't sound like a CPU problem then.&lt;BR /&gt;&lt;BR /&gt;Could it be related to Oracle doing a log switch and or checkpoint? This can cause a lot of disk activity. Also it's not a good idea having your Oracle redo log files on Autoraid.&lt;BR /&gt;&lt;BR /&gt;Check your Oracle alert log for log switches.&lt;BR /&gt;&lt;BR /&gt;I'd also concentrate on disk I/O with glance (d u and i screens).&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;John&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 05 Oct 2000 11:10:08 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450515#M10811</guid>
      <dc:creator>John Palmer</dc:creator>
      <dc:date>2000-10-05T11:10:08Z</dc:date>
    </item>
    <item>
      <title>Re: Benchmark tests</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450516#M10812</link>
      <description>Could it be that some transactions are larger than others hence take longer to process?&lt;BR /&gt;&lt;BR /&gt;Can you monitor ORACLE , get a snap shot every 5 minutes or so and see what is being processed , see the size of the records etc..</description>
      <pubDate>Thu, 05 Oct 2000 11:12:10 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/benchmark-tests/m-p/2450516#M10812</guid>
      <dc:creator>Devbinder Singh Marway</dc:creator>
      <dc:date>2000-10-05T11:12:10Z</dc:date>
    </item>
  </channel>
</rss>

