<?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: Attn: Peter Jansen in Disk Enclosures</title>
    <link>https://community.hpe.com/t5/disk-enclosures/attn-peter-jansen/m-p/2487729#M1085</link>
    <description>Hi Peter,&lt;BR /&gt;the Write Cache and Read Cache enable is a bit which have nothing to do with the cache of the diskarray itself. These bit are for the Host interface which could set there interal performance to a higher level. These could be a performance inrease because of the Host interface setting.&lt;BR /&gt;&lt;BR /&gt;Best regards&lt;BR /&gt;&lt;BR /&gt;Frank Skibbe</description>
    <pubDate>Tue, 30 Jan 2001 06:34:56 GMT</pubDate>
    <dc:creator>Frank Skibbe</dc:creator>
    <dc:date>2001-01-30T06:34:56Z</dc:date>
    <item>
      <title>Attn: Peter Jansen</title>
      <link>https://community.hpe.com/t5/disk-enclosures/attn-peter-jansen/m-p/2487728#M1084</link>
      <description>Hi Peter -&lt;BR /&gt;&lt;BR /&gt;I was reading your response to a question regard performance problems on a autoraid 12H array (see below).&lt;BR /&gt;&lt;BR /&gt;Would you mind expanding on this a bit?  How could it be a performance problem to have the cache enabled?  I just bought an autoraid and had dual 96mb controllers installed specifically so I could take advantage of the cache.&lt;BR /&gt;&lt;BR /&gt;Thanks in advance!&lt;BR /&gt;Tim&lt;BR /&gt;timmed@saif.com&lt;BR /&gt;&lt;BR /&gt;----------------------------------&lt;BR /&gt;&lt;BR /&gt;Another big performancehit can be suffered if the AutoRaid "automatically" sets the "Write Cache" to "Enabled". &lt;BR /&gt;&lt;BR /&gt;Please check if this is true for your AutoRaid (i think the command is &lt;ARRAYDSP -a=""&gt;. If so, please let me know to supply you with further information on this subject. &lt;BR /&gt;&lt;BR /&gt;Best regards, &lt;BR /&gt;&lt;BR /&gt;Peter Jansen &lt;BR /&gt;ITchallenge&lt;/ARRAYDSP&gt;</description>
      <pubDate>Mon, 29 Jan 2001 21:29:03 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/attn-peter-jansen/m-p/2487728#M1084</guid>
      <dc:creator>Tim Medford</dc:creator>
      <dc:date>2001-01-29T21:29:03Z</dc:date>
    </item>
    <item>
      <title>Re: Attn: Peter Jansen</title>
      <link>https://community.hpe.com/t5/disk-enclosures/attn-peter-jansen/m-p/2487729#M1085</link>
      <description>Hi Peter,&lt;BR /&gt;the Write Cache and Read Cache enable is a bit which have nothing to do with the cache of the diskarray itself. These bit are for the Host interface which could set there interal performance to a higher level. These could be a performance inrease because of the Host interface setting.&lt;BR /&gt;&lt;BR /&gt;Best regards&lt;BR /&gt;&lt;BR /&gt;Frank Skibbe</description>
      <pubDate>Tue, 30 Jan 2001 06:34:56 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/attn-peter-jansen/m-p/2487729#M1085</guid>
      <dc:creator>Frank Skibbe</dc:creator>
      <dc:date>2001-01-30T06:34:56Z</dc:date>
    </item>
    <item>
      <title>Re: Attn: Peter Jansen</title>
      <link>https://community.hpe.com/t5/disk-enclosures/attn-peter-jansen/m-p/2487730#M1086</link>
      <description>Gents,&lt;BR /&gt;&lt;BR /&gt;Some explanation about the READ &amp;amp; Write Cache settings:&lt;BR /&gt;Change the apparent state of write cache on or&lt;BR /&gt;off. This switch only changes the write cache&lt;BR /&gt;setting presented to the operating system and does not affect the operation of the disk array, which always has write cache enabled. The disk array write cache is stored in NVRAM which eliminates the need to disable write cache for protection against power loss. This switch is provided for interoperability with those operating systems that require write cache be turned off.&lt;BR /&gt;However arraymgr -w off $ARRAYID will return an error message to you, you are not able to change this setting (allthough the man page says so).&lt;BR /&gt;The setting of the write cache is directly related to the resiliency mode settings and therefore cannot be changed manually with the&lt;BR /&gt;'arraymgr -w off' command. The write cache flag setting is as follows:&lt;BR /&gt; - Resiliency Setting = SingleController -&amp;gt; Write Cache = DISABLED&lt;BR /&gt;   Force Unit Access Response   = 2&lt;BR /&gt; - Resiliency Setting = Secure           -&amp;gt; Write Cache = DISABLED&lt;BR /&gt;   Force Unit Access Response   = 2&lt;BR /&gt; - Resiliency Setting = Normal           -&amp;gt; Write Cache = ENABLED&lt;BR /&gt;   Force Unit Access Response   = 2&lt;BR /&gt; - Resiliency Setting = HighPerformance  -&amp;gt; Write Cache = ENABLED&lt;BR /&gt;   Force Unit Access Response   = 0&lt;BR /&gt;&lt;BR /&gt;The 'Force Unit Access Response' parameter depends on the resiliency mode settings. In 'SingleController', 'Normal' and 'Secure' mode the FUA parameter is 2. In 'HighPerformance' mode the FUA parameter is 0.&lt;BR /&gt;The FUA parameter controls the array's response to the FUA bit. The FUA bit is a command from the host when Write Cache is enabled. It gives the host an opportunity to flush write cache on command:&lt;BR /&gt;&lt;BR /&gt; - if this field is 0, then the FUA bit is ignored.&lt;BR /&gt; - If this field is 1, then the FUA bit in a write command disables immediate report on the write and all write cache data for the involved LUN will be flushed.&lt;BR /&gt; - If this field is 2, then the FUA bit in a write command disables immediate report on the write and all write cache data for the involved LUN will be flushed along with the   map journal before completing the write request.&lt;BR /&gt; - If NVRAM use is disabled due to the chosen map resiliency mode, then this field is ignored.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;&lt;BR /&gt;Mark</description>
      <pubDate>Tue, 30 Jan 2001 13:57:23 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/attn-peter-jansen/m-p/2487730#M1086</guid>
      <dc:creator>Mark van Silfhout</dc:creator>
      <dc:date>2001-01-30T13:57:23Z</dc:date>
    </item>
  </channel>
</rss>

