<?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: eva + lvm striping in Disk Enclosures</title>
    <link>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621473#M17501</link>
    <description>If you have free resources, then test the different configurations, to the real impact.&lt;BR /&gt;Mike</description>
    <pubDate>Fri, 09 Sep 2005 07:04:07 GMT</pubDate>
    <dc:creator>Mike Reznak</dc:creator>
    <dc:date>2005-09-09T07:04:07Z</dc:date>
    <item>
      <title>eva + lvm striping</title>
      <link>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621468#M17496</link>
      <description>i would just like to know if there is any benefit in implementing this:&lt;BR /&gt;&lt;BR /&gt;eva with a few virtual disk and then lvm stripe them into a single vg.&lt;BR /&gt;&lt;BR /&gt;will there be performance improvement? is it a best practice?</description>
      <pubDate>Thu, 08 Sep 2005 04:57:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621468#M17496</guid>
      <dc:creator>ChewSP</dc:creator>
      <dc:date>2005-09-08T04:57:05Z</dc:date>
    </item>
    <item>
      <title>Re: eva + lvm striping</title>
      <link>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621469#M17497</link>
      <description>Hi,&lt;BR /&gt;No, it will not give any performance benefit nor be any "best practice".&lt;BR /&gt;&lt;BR /&gt;You can activate the full performance of the EVA, using one single LUN.</description>
      <pubDate>Thu, 08 Sep 2005 04:59:55 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621469#M17497</guid>
      <dc:creator>Leif Halvarsson_2</dc:creator>
      <dc:date>2005-09-08T04:59:55Z</dc:date>
    </item>
    <item>
      <title>Re: eva + lvm striping</title>
      <link>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621470#M17498</link>
      <description>Be aware of this:&lt;BR /&gt;&lt;BR /&gt;When presenting only one vdsik to a HP-UX System you have to change the kernel paramater:&lt;BR /&gt;scsi_max_depth - standard is 8&lt;BR /&gt;&lt;BR /&gt;Some of our customers had very poor performance, aftre raising this paramter everything works fine.&lt;BR /&gt;&lt;BR /&gt;This parameter must be set in VMware ESX environments too, when using only one LUN.&lt;BR /&gt;&lt;BR /&gt;Markus</description>
      <pubDate>Thu, 08 Sep 2005 06:00:12 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621470#M17498</guid>
      <dc:creator>Markus_10</dc:creator>
      <dc:date>2005-09-08T06:00:12Z</dc:date>
    </item>
    <item>
      <title>Re: eva + lvm striping</title>
      <link>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621471#M17499</link>
      <description>Look at this:&lt;BR /&gt;&lt;A href="http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=lpg35039" target="_blank"&gt;http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=lpg35039&lt;/A&gt; &lt;BR /&gt;&lt;BR /&gt;And Leif: In order to achieve full performance of an EVA you need to allocate at least 2 LUNs, 1 on each controller!&lt;BR /&gt;&lt;BR /&gt;Cheers&lt;BR /&gt;Peter</description>
      <pubDate>Thu, 08 Sep 2005 08:48:37 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621471#M17499</guid>
      <dc:creator>Peter Mattei</dc:creator>
      <dc:date>2005-09-08T08:48:37Z</dc:date>
    </item>
    <item>
      <title>Re: eva + lvm striping</title>
      <link>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621472#M17500</link>
      <description>I don't think its a good idea.  There would be little or no performance gain with LVM striping. &lt;BR /&gt;&lt;BR /&gt;As you might be aware , EVA does the striping of its own. So there would be double striping .If you enable LVM striping , there are overheads involved as well. From OS perception , there would be slight LVM striping overhead . From EVA perception, it is little more overhead to it as you are making him work more. Although EVA is powerful enough to handle the overhead, it is always good to allow him to work of its own. &lt;BR /&gt;&lt;BR /&gt;To get maximum throuput from EVA you can have 1 disk group and Vdisks/LUNs equally distributed/balanced on the two controllers.&lt;BR /&gt;&lt;BR /&gt;Although it is not mentioned in HP's best practices about LVM striping, I would say it is not required.</description>
      <pubDate>Thu, 08 Sep 2005 17:18:17 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621472#M17500</guid>
      <dc:creator>Sameer_Nirmal</dc:creator>
      <dc:date>2005-09-08T17:18:17Z</dc:date>
    </item>
    <item>
      <title>Re: eva + lvm striping</title>
      <link>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621473#M17501</link>
      <description>If you have free resources, then test the different configurations, to the real impact.&lt;BR /&gt;Mike</description>
      <pubDate>Fri, 09 Sep 2005 07:04:07 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621473#M17501</guid>
      <dc:creator>Mike Reznak</dc:creator>
      <dc:date>2005-09-09T07:04:07Z</dc:date>
    </item>
    <item>
      <title>Re: eva + lvm striping</title>
      <link>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621474#M17502</link>
      <description>Sorry. Once more the whole sentence...&lt;BR /&gt;If you have free resources, then test the different configurations, to see the real impact.&lt;BR /&gt;Mike</description>
      <pubDate>Fri, 09 Sep 2005 07:05:17 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/eva-lvm-striping/m-p/3621474#M17502</guid>
      <dc:creator>Mike Reznak</dc:creator>
      <dc:date>2005-09-09T07:05:17Z</dc:date>
    </item>
  </channel>
</rss>

