<?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: Question on VA in Disk Enclosures</title>
    <link>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784258#M5744</link>
    <description>The classic system activity reporter.....&lt;BR /&gt;&lt;BR /&gt;"sar -d &lt;TIME&gt; &lt;N&gt;"&lt;BR /&gt;&lt;BR /&gt;time &amp;gt; 60 (seconds)&lt;BR /&gt;n - number of times you should repeat sample.&lt;BR /&gt;&lt;BR /&gt;so sar -d 15 2&lt;BR /&gt;It will do two samples of 15 seconds.  In this the last two columns are interesting as it gives Average Service time (or slope of disk % Vs IO/s).  This is the average time for an IO to be processed, also how long it took for the IO to hit the disk subsystem.&lt;BR /&gt;&lt;BR /&gt;The problem with sar is that&lt;BR /&gt; o It is not as accurate as MW &lt;BR /&gt; o Is that it is not historical - However, you can set it up using cron &amp;amp; sa&lt;BR /&gt;&lt;BR /&gt;I'm a BIG fan of MeasureWare and have solved / helped solve many a problem with it&lt;BR /&gt;&lt;BR /&gt;Regards&lt;BR /&gt;&lt;BR /&gt;Tim&lt;/N&gt;&lt;/TIME&gt;</description>
    <pubDate>Tue, 13 Aug 2002 17:29:27 GMT</pubDate>
    <dc:creator>Tim D Fulford</dc:creator>
    <dc:date>2002-08-13T17:29:27Z</dc:date>
    <item>
      <title>Question on VA</title>
      <link>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784252#M5738</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;We have purchase the new L2000 with VA7100 and would like to know couple things.  What is the throughput of the new box's I/O subsystem?  I don't mean how fast is the fibre channel, rpms of the disks, etc.  What I really would like is something like a megs/sec of random read/write consistent megs/sec of random read/write burst.  I say random because that???s mostly how our application system works (as opposed to sequential).&lt;BR /&gt;&lt;BR /&gt;If anyone has any links or docs, I will appreciate that or even answer to my questions.&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;&lt;BR /&gt;Mike</description>
      <pubDate>Mon, 12 Aug 2002 15:16:31 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784252#M5738</guid>
      <dc:creator>Mike_305</dc:creator>
      <dc:date>2002-08-12T15:16:31Z</dc:date>
    </item>
    <item>
      <title>Re: Question on VA</title>
      <link>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784253#M5739</link>
      <description>Hello,&lt;BR /&gt;&lt;BR /&gt;Is this what you are looking for?&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://www.hp.com/products1/storage/products/disk_arrays/midrange/va7100/infolibrary/7100_product_brief.pdf" target="_blank"&gt;http://www.hp.com/products1/storage/products/disk_arrays/midrange/va7100/infolibrary/7100_product_brief.pdf&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://www.hp.com/products1/storage/products/disk_arrays/infolibrary/va_perf_analysis.pdf" target="_blank"&gt;http://www.hp.com/products1/storage/products/disk_arrays/infolibrary/va_perf_analysis.pdf&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;Hope this helps&lt;BR /&gt;&lt;BR /&gt;Chris</description>
      <pubDate>Mon, 12 Aug 2002 15:23:50 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784253#M5739</guid>
      <dc:creator>Christopher McCray_1</dc:creator>
      <dc:date>2002-08-12T15:23:50Z</dc:date>
    </item>
    <item>
      <title>Re: Question on VA</title>
      <link>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784254#M5740</link>
      <description>Hi Mike:&lt;BR /&gt;&lt;BR /&gt;I dont know if this is what your after, but this link is nice for getting some information.&lt;BR /&gt;&lt;A href="http://www.spec.org/" target="_blank"&gt;http://www.spec.org/&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;on the left side you can select search, then &lt;BR /&gt;you can chose like system, and hp, and it will give you a listing and you can select which model you want to see benchmarks for.&lt;BR /&gt;&lt;BR /&gt;scott&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Mon, 12 Aug 2002 17:57:58 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784254#M5740</guid>
      <dc:creator>Scott_14</dc:creator>
      <dc:date>2002-08-12T17:57:58Z</dc:date>
    </item>
    <item>
      <title>Re: Question on VA</title>
      <link>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784255#M5741</link>
      <description>Hi&lt;BR /&gt;When I looked at the VA7100 and VA7400 some month ago I could find performance specifications in the documentation but now I can't longer find it. If I remember correct it was something like&lt;BR /&gt;&lt;BR /&gt;VA7100: 3500 IOPS disk, 15000 IOPS chache, 90MB/s transfer rate. &lt;BR /&gt;VA7400: About double compared to VA7100.&lt;BR /&gt;Have a look at:&lt;BR /&gt;&lt;A href="http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0x7239eea29889d611abdb0090277a778c,00.html" target="_blank"&gt;http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0x7239eea29889d611abdb0090277a778c,00.html&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;There you can find a file system benchmark and some test results.</description>
      <pubDate>Mon, 12 Aug 2002 18:48:50 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784255#M5741</guid>
      <dc:creator>Leif Halvarsson_2</dc:creator>
      <dc:date>2002-08-12T18:48:50Z</dc:date>
    </item>
    <item>
      <title>Re: Question on VA</title>
      <link>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784256#M5742</link>
      <description>OK&lt;BR /&gt;&lt;BR /&gt;I'm going to answer this in a weird way, hang on...&lt;BR /&gt;&lt;BR /&gt;We had fc60 two controllers, 12 LUNS, 256MB cache per controller.  We used this for a OLTP DB(Random) and we &lt;BR /&gt;I used to plot graphs, and I would get a pritty straight line of disk % Vs IO/s.  The gradient of the line gives the average service time, 4-5ms.  Pritty good, each IO was about 2-4kB so...max throughput&lt;BR /&gt;12LUNx3kB(1/0.0045) ==&amp;gt; 8,000kB/s or 8MB/s (could see more)&lt;BR /&gt;&lt;BR /&gt;We have now replaced the fc60 with VA7400, two controllers 512MB per controller.  I see same behaviour except service times are 1-2ms.  Thus...&lt;BR /&gt;12LUNx3kBx(1/0.0015) ==&amp;gt; 24,000kB/s&lt;BR /&gt;&lt;BR /&gt;The VA7100 is a "lesser" disk-subsystem. but I think it has similar CPU's AND it can hold LOTS of disks than an fc60.  You can increase the random throughput simply by striping over more LUNS.  Also remember that bigger disks can give performance gains as well as faster disks as the head does not have to travel so far between tracks.&lt;BR /&gt;&lt;BR /&gt;You really need to test it against your database/Application to get a good figure, but If you already know what the demands are now; on your current system; you have a head start.  I'd start by plotting disk% Vs IO/s &amp;amp; see what that looks like (MeasureWare)&lt;BR /&gt;&lt;BR /&gt;Regards&lt;BR /&gt;&lt;BR /&gt;Tim</description>
      <pubDate>Mon, 12 Aug 2002 20:17:28 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784256#M5742</guid>
      <dc:creator>Tim D Fulford</dc:creator>
      <dc:date>2002-08-12T20:17:28Z</dc:date>
    </item>
    <item>
      <title>Re: Question on VA</title>
      <link>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784257#M5743</link>
      <description>Hi Tim,&lt;BR /&gt;&lt;BR /&gt;I know at the bottom of your e-mail you told Measure Ware to use, do you have any other recommendation for monitoring performance or monitoring IO.&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;&lt;BR /&gt;Mike&lt;BR /&gt;</description>
      <pubDate>Mon, 12 Aug 2002 21:23:39 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784257#M5743</guid>
      <dc:creator>Mike_305</dc:creator>
      <dc:date>2002-08-12T21:23:39Z</dc:date>
    </item>
    <item>
      <title>Re: Question on VA</title>
      <link>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784258#M5744</link>
      <description>The classic system activity reporter.....&lt;BR /&gt;&lt;BR /&gt;"sar -d &lt;TIME&gt; &lt;N&gt;"&lt;BR /&gt;&lt;BR /&gt;time &amp;gt; 60 (seconds)&lt;BR /&gt;n - number of times you should repeat sample.&lt;BR /&gt;&lt;BR /&gt;so sar -d 15 2&lt;BR /&gt;It will do two samples of 15 seconds.  In this the last two columns are interesting as it gives Average Service time (or slope of disk % Vs IO/s).  This is the average time for an IO to be processed, also how long it took for the IO to hit the disk subsystem.&lt;BR /&gt;&lt;BR /&gt;The problem with sar is that&lt;BR /&gt; o It is not as accurate as MW &lt;BR /&gt; o Is that it is not historical - However, you can set it up using cron &amp;amp; sa&lt;BR /&gt;&lt;BR /&gt;I'm a BIG fan of MeasureWare and have solved / helped solve many a problem with it&lt;BR /&gt;&lt;BR /&gt;Regards&lt;BR /&gt;&lt;BR /&gt;Tim&lt;/N&gt;&lt;/TIME&gt;</description>
      <pubDate>Tue, 13 Aug 2002 17:29:27 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/question-on-va/m-p/2784258#M5744</guid>
      <dc:creator>Tim D Fulford</dc:creator>
      <dc:date>2002-08-13T17:29:27Z</dc:date>
    </item>
  </channel>
</rss>

