<?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: MSA 60 + Linux + SATA setup for performance in Disk Enclosures</title>
    <link>https://community.hpe.com/t5/disk-enclosures/msa-60-linux-sata-setup-for-performance/m-p/4667395#M35564</link>
    <description>New test result.&lt;BR /&gt;&lt;BR /&gt;We noticed that the card itself was the problem.&lt;BR /&gt;By testing the similare things using the motherboards built in card, we noticed that the load did not droped off, if i/o was added from a different card.&lt;BR /&gt;&lt;BR /&gt;</description>
    <pubDate>Thu, 29 Jul 2010 16:59:00 GMT</pubDate>
    <dc:creator>Tomas Dalebjörk</dc:creator>
    <dc:date>2010-07-29T16:59:00Z</dc:date>
    <item>
      <title>MSA 60 + Linux + SATA setup for performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/msa-60-linux-sata-setup-for-performance/m-p/4667394#M35563</link>
      <description>Hi,&lt;BR /&gt;We are trying to setup the SATA array for the following performance.&lt;BR /&gt;Our need: 200MB/s write with 256k block&lt;BR /&gt;At same time 200MB/s read with 256k block&lt;BR /&gt;&lt;BR /&gt;We have today baught:&lt;BR /&gt;12x1TB SATA disks&lt;BR /&gt;and we would like to have 10 of thoose disk shelfs.&lt;BR /&gt;&lt;BR /&gt;We have seen the below performance:&lt;BR /&gt;250MB/s write of 10 streams, with 256k block&lt;BR /&gt;130MB/s read from 10 streams, with 256k block&lt;BR /&gt;&lt;BR /&gt;But when we combining 5 reads and 5 writes, the read performance drops down to 25MB/s&lt;BR /&gt;and write performance drops down to 75MB/s&lt;BR /&gt;&lt;BR /&gt;Adding a new disk shelf, and some load on that disk shelf, resulted in equal load on each disk shelf, but only 50% of the above figures.!?&lt;BR /&gt;&lt;BR /&gt;The disk shelfs are cascaded from one and same controller card, and same port.&lt;BR /&gt;&lt;BR /&gt;I even tried to test this from another disk shelf, which was connected to another port, but still on same controller card, but with same result: 50% on each disk shelf, and with performance drop off.&lt;BR /&gt;&lt;BR /&gt;Any one have an idea how to improve this?&lt;BR /&gt;And how to setup the HP server so that it could scale for performance.&lt;BR /&gt;&lt;BR /&gt;Please contact me.&lt;BR /&gt;&lt;BR /&gt;Regards Tomas</description>
      <pubDate>Wed, 28 Jul 2010 19:27:12 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/msa-60-linux-sata-setup-for-performance/m-p/4667394#M35563</guid>
      <dc:creator>Tomas Dalebjörk</dc:creator>
      <dc:date>2010-07-28T19:27:12Z</dc:date>
    </item>
    <item>
      <title>Re: MSA 60 + Linux + SATA setup for performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/msa-60-linux-sata-setup-for-performance/m-p/4667395#M35564</link>
      <description>New test result.&lt;BR /&gt;&lt;BR /&gt;We noticed that the card itself was the problem.&lt;BR /&gt;By testing the similare things using the motherboards built in card, we noticed that the load did not droped off, if i/o was added from a different card.&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 29 Jul 2010 16:59:00 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/msa-60-linux-sata-setup-for-performance/m-p/4667395#M35564</guid>
      <dc:creator>Tomas Dalebjörk</dc:creator>
      <dc:date>2010-07-29T16:59:00Z</dc:date>
    </item>
  </channel>
</rss>

