<?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: Flow control: What are your real world experiences? in Array Performance and Data Protection</title>
    <link>https://community.hpe.com/t5/array-performance-and-data/flow-control-what-are-your-real-world-experiences/m-p/6985331#M861</link>
    <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Hi Dan,&lt;/P&gt;&lt;P&gt;I've seen issues in a SmartStack when flow control hasn't been enabled. During high performance load and a manual controller failover, we observed 2 things:&lt;/P&gt;&lt;P&gt;1) the time for activity to resume after the failover took longer than with flow control enabled&lt;/P&gt;&lt;P&gt;2) 'choppy' performance (peaks and troughs in monitor&amp;gt;performance charts for IOPS) on the Nimble rather than consistent performance &lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
    <pubDate>Tue, 30 Jun 2015 09:12:23 GMT</pubDate>
    <dc:creator>amirul93</dc:creator>
    <dc:date>2015-06-30T09:12:23Z</dc:date>
    <item>
      <title>Flow control: What are your real world experiences?</title>
      <link>https://community.hpe.com/t5/array-performance-and-data/flow-control-what-are-your-real-world-experiences/m-p/6985330#M860</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;We have an internal discussion going that revolves around flow control.&lt;/P&gt;&lt;P&gt;Arguments for enabling it:&lt;/P&gt;&lt;P&gt;-Most every storage vendor recommends enabling it.&lt;/P&gt;&lt;P&gt;-It can be more efficient at managing traffic than TCP dropping and re-sending packets.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Arguments against enabling it:&lt;/P&gt;&lt;P&gt;-TCP was designed to handle its own flows.&lt;/P&gt;&lt;P&gt;-Pause frames from a host unable to handle traffic can negatively impact other hosts currently not having issues.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Anyone have any real world experiences where flow control either improved or hindered performance?&lt;/P&gt;&lt;P&gt;Any other insight?&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Thu, 04 Jun 2015 16:21:27 GMT</pubDate>
      <guid>https://community.hpe.com/t5/array-performance-and-data/flow-control-what-are-your-real-world-experiences/m-p/6985330#M860</guid>
      <dc:creator>foodandbikes</dc:creator>
      <dc:date>2015-06-04T16:21:27Z</dc:date>
    </item>
    <item>
      <title>Re: Flow control: What are your real world experiences?</title>
      <link>https://community.hpe.com/t5/array-performance-and-data/flow-control-what-are-your-real-world-experiences/m-p/6985331#M861</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Hi Dan,&lt;/P&gt;&lt;P&gt;I've seen issues in a SmartStack when flow control hasn't been enabled. During high performance load and a manual controller failover, we observed 2 things:&lt;/P&gt;&lt;P&gt;1) the time for activity to resume after the failover took longer than with flow control enabled&lt;/P&gt;&lt;P&gt;2) 'choppy' performance (peaks and troughs in monitor&amp;gt;performance charts for IOPS) on the Nimble rather than consistent performance &lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Tue, 30 Jun 2015 09:12:23 GMT</pubDate>
      <guid>https://community.hpe.com/t5/array-performance-and-data/flow-control-what-are-your-real-world-experiences/m-p/6985331#M861</guid>
      <dc:creator>amirul93</dc:creator>
      <dc:date>2015-06-30T09:12:23Z</dc:date>
    </item>
  </channel>
</rss>

