<?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 stress test in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511080#M894908</link>
    <description>Hi,&lt;BR /&gt;&lt;BR /&gt;Any body have idea how can I get benchmark software , or other software that can generate load on machine ?&lt;BR /&gt;&lt;BR /&gt;thanks in advance,</description>
    <pubDate>Thu, 29 Mar 2001 08:50:50 GMT</pubDate>
    <dc:creator>Printaporn_1</dc:creator>
    <dc:date>2001-03-29T08:50:50Z</dc:date>
    <item>
      <title>stress test</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511080#M894908</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;Any body have idea how can I get benchmark software , or other software that can generate load on machine ?&lt;BR /&gt;&lt;BR /&gt;thanks in advance,</description>
      <pubDate>Thu, 29 Mar 2001 08:50:50 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511080#M894908</guid>
      <dc:creator>Printaporn_1</dc:creator>
      <dc:date>2001-03-29T08:50:50Z</dc:date>
    </item>
    <item>
      <title>Re: stress test</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511081#M894910</link>
      <description>&lt;BR /&gt;You dont really need software to do this. There is one simple command which will stress test any box as much as you want. We use it all the time; dd&lt;BR /&gt;&lt;BR /&gt;If you simply kick off as many dd commands at the same time as you want until the load average gets to a nice high level this is sufficient to really stress test your box.&lt;BR /&gt;&lt;BR /&gt;dd if=/dev/rdsk/cxxxxx of=/dev/null bs=1024k &amp;amp;&lt;BR /&gt;&lt;BR /&gt;On a K class we would kick off one for each disk we had on the server (which was around 60), this would run the load average at around 100 which would stress the box nicely. In the script when the first dd finished we would kick off another so we always had the same heavy load. We would ideally run this for a week on a new server. Usually in that time we would find at least 1 disk or controller which would fail, but thats what stress testing is for!&lt;BR /&gt;</description>
      <pubDate>Thu, 29 Mar 2001 09:01:20 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511081#M894910</guid>
      <dc:creator>Edward Sedgemore</dc:creator>
      <dc:date>2001-03-29T09:01:20Z</dc:date>
    </item>
    <item>
      <title>Re: stress test</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511082#M894911</link>
      <description>Hi!&lt;BR /&gt;&lt;BR /&gt;For a simple floating point performance test I use a matrix multiplication program. With an increasing matrix size the effects of the cache of the processor can be reduced. Such a program is also good to generate processor and memory load.&lt;BR /&gt;&lt;BR /&gt;Hope this helps, Thomas</description>
      <pubDate>Thu, 29 Mar 2001 09:08:20 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511082#M894911</guid>
      <dc:creator>Thomas Kollig</dc:creator>
      <dc:date>2001-03-29T09:08:20Z</dc:date>
    </item>
    <item>
      <title>Re: stress test</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511083#M894913</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;With respect to I/O benchmarking, stkio is a very good tool which you can download from the storagetek website.&lt;BR /&gt;&lt;BR /&gt;It allows you to specify numerous parameters such as block size and test on specific filesystems or raw devices the I/O throughput for random reads and writes as well as sequential reads and writes. It presents the results in MB/s, IO/s as well as total time taken in ms.&lt;BR /&gt;&lt;BR /&gt;For commercial load-generation software, you can take a look at Mercury Interactive.&lt;BR /&gt;&lt;BR /&gt;Hope this helps. Regards.&lt;BR /&gt;&lt;BR /&gt;Steven Sim Kok Leong&lt;BR /&gt;Brainbench MVP for Unix Admin&lt;BR /&gt;&lt;A href="http://www.brainbench.com" target="_blank"&gt;http://www.brainbench.com&lt;/A&gt;</description>
      <pubDate>Thu, 29 Mar 2001 09:09:22 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511083#M894913</guid>
      <dc:creator>Steven Sim Kok Leong</dc:creator>
      <dc:date>2001-03-29T09:09:22Z</dc:date>
    </item>
    <item>
      <title>Re: stress test</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511084#M894914</link>
      <description>I personally felt that program compiling is one of the better way to stress test a system.  For example, a server with a faulty CPU may lost for a couple days before it reboots itself.  But, if you get it to compile samba, I bet you the server will be rebooted within 10 minutes.&lt;BR /&gt;&lt;BR /&gt;~Philip</description>
      <pubDate>Thu, 29 Mar 2001 09:22:23 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/stress-test/m-p/2511084#M894914</guid>
      <dc:creator>Philip Chan_1</dc:creator>
      <dc:date>2001-03-29T09:22:23Z</dc:date>
    </item>
  </channel>
</rss>

