<?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: VA7400 performance in Disk Enclosures</title>
    <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757902#M5347</link>
    <description>Hi&lt;BR /&gt;I agree, there must be something wrong, perhaps firmware upgrading will help.&lt;BR /&gt;&lt;BR /&gt;My test was from a CMD 7400 RAID controller and a RAID 5 set with only 4 drives. A VA7400 should perform much better.</description>
    <pubDate>Fri, 05 Jul 2002 11:08:54 GMT</pubDate>
    <dc:creator>Leif Halvarsson_2</dc:creator>
    <dc:date>2002-07-05T11:08:54Z</dc:date>
    <item>
      <title>VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757895#M5340</link>
      <description>I have a VA7400 array connected to a RP7400 by two Brocade switches.&lt;BR /&gt;The system is esentially an Oracle machine.&lt;BR /&gt;Using Glance, I can see low CPU utilization (20-60%) but frequent disk bottlenecks (i.e high system load).&lt;BR /&gt;I already read forum postings and HP docs about performance issues, I put array in High Performance mode, LUNs in RG1 use Controller 1 as primary path (and viceversa), etc.&lt;BR /&gt;Using CommandView metrics, I can see a Total I/O throughput lower than 40 units/sec.&lt;BR /&gt;&lt;BR /&gt;Do you think that putting all Oracle file on the same, big LUN can cause this performance issues?&lt;BR /&gt;A firmware upgrade (actually I have HP13) can also be useful?&lt;BR /&gt;Other suggestions?&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 04 Jul 2002 11:58:43 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757895#M5340</guid>
      <dc:creator>Domenico Viggiani</dc:creator>
      <dc:date>2002-07-04T11:58:43Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757896#M5341</link>
      <description>Hi&lt;BR /&gt;I can said you that the last release of FW could for sure solve some performance issues&lt;BR /&gt;&lt;BR /&gt;bye&lt;BR /&gt;&lt;BR /&gt;marino</description>
      <pubDate>Thu, 04 Jul 2002 12:43:58 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757896#M5341</guid>
      <dc:creator>Marino Meloni_1</dc:creator>
      <dc:date>2002-07-04T12:43:58Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757897#M5342</link>
      <description>How many disks are in the array, what type? The VA7400 performs best with &amp;gt;45 disks installed!&lt;BR /&gt;What mode are you running? &lt;BR /&gt;AutoRaid or RAID01?&lt;BR /&gt;If AutoRaid, what is the R01/R5DP ratio?&lt;BR /&gt;For OLTP and high write IO rate Raid01 performs much better. This is not an error just normal RAID5 behaviour (RAID5 write penalty).&lt;BR /&gt;&lt;BR /&gt;Peter</description>
      <pubDate>Fri, 05 Jul 2002 06:44:36 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757897#M5342</guid>
      <dc:creator>Peter Mattei</dc:creator>
      <dc:date>2002-07-05T06:44:36Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757898#M5343</link>
      <description>Hi&lt;BR /&gt;I dont think using one or more LUNs make any performance difference.&lt;BR /&gt;&lt;BR /&gt;If you suspect bad file system performance you can use the postmark benchmark (attached) for checking. &lt;BR /&gt;Compile it (cc postmark.c).&lt;BR /&gt;Rename the output file (mv a.out postmark).&lt;BR /&gt;Move the file to a directory on the filesystem you want to test (mv postmark /xxx/yyy/zzz)&lt;BR /&gt;Start postmark (./postmark)&lt;BR /&gt;At the pm&amp;gt; prompt:&lt;BR /&gt;Set parameters (pm&amp;gt; set number 30000, pm&amp;gt; set transactions 5000)&lt;BR /&gt;You can test with different values.&lt;BR /&gt;Run the test (pm&amp;gt; run)&lt;BR /&gt;&lt;BR /&gt;Here is the output from my RAID system.&lt;BR /&gt;&lt;BR /&gt;Time:&lt;BR /&gt;        147 seconds total&lt;BR /&gt;        34 seconds of transactions (147 per second)&lt;BR /&gt;&lt;BR /&gt;Files:&lt;BR /&gt;        32453 created (220 per second)&lt;BR /&gt;                Creation alone: 30000 files (384 per second)&lt;BR /&gt;                Mixed with transactions: 2453 files (72 per second)&lt;BR /&gt;        2532 read (74 per second)&lt;BR /&gt;        2468 appended (72 per second)&lt;BR /&gt;        32453 deleted (220 per second)&lt;BR /&gt;                Deletion alone: 29906 files (854 per second)&lt;BR /&gt;                Mixed with transactions: 2547 files (74 per second)&lt;BR /&gt;&lt;BR /&gt;Data:&lt;BR /&gt;        12.67 megabytes read (88.24 kilobytes per second)&lt;BR /&gt;        167.31 megabytes written (1.14 megabytes per second)&lt;BR /&gt;&lt;BR /&gt;With a VA 7400 I think you will get better performance then me.</description>
      <pubDate>Fri, 05 Jul 2002 08:28:42 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757898#M5343</guid>
      <dc:creator>Leif Halvarsson_2</dc:creator>
      <dc:date>2002-07-05T08:28:42Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757899#M5344</link>
      <description>I think you should put your rollback segments on a 2nd LUN, also put one set of control files on this 2nd LUN.&lt;BR /&gt;Some people recommend putting indexes in a separate LUN from data, you might try this but I think it depends on your own experience of the application and tablespace usage.  For instance if you have a high i/o tablespace then it may be worth moving it away from the others.&lt;BR /&gt;Route the traffic to this 2nd LUN through the other switch.&lt;BR /&gt;&lt;BR /&gt;I personally like to keep about 4 LUNS for Oracle and I have noticed that it does seem allocate separate LUNS to separate disks, so it does make a difference.&lt;BR /&gt;&lt;BR /&gt;Make sure that your waits are Oracle i/o based and not caused by semaphores or (even worse) memory paging.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 05 Jul 2002 09:12:48 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757899#M5344</guid>
      <dc:creator>Steve Lewis</dc:creator>
      <dc:date>2002-07-05T09:12:48Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757900#M5345</link>
      <description>I read also other messaages about VA7100, all suggesting the upgrade to HP16 firmware.&lt;BR /&gt;&lt;BR /&gt;Actually array has 14*3=42 disks (36 GB) but installation of others 2*15 = 30 disks (72 GB) is in progress.&lt;BR /&gt;&lt;BR /&gt;Array is configured in AutoRAID mode (I don't know R01/R5DP ratio).&lt;BR /&gt;&lt;BR /&gt;I will try to run postmark benchmark next week. Thank you very much, Leif!&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 05 Jul 2002 10:18:15 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757900#M5345</guid>
      <dc:creator>Domenico Viggiani</dc:creator>
      <dc:date>2002-07-05T10:18:15Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757901#M5346</link>
      <description>These are my numbers:&lt;BR /&gt;&lt;BR /&gt;Time:&lt;BR /&gt;        355 seconds total&lt;BR /&gt;        146 seconds of transactions (34 per second)&lt;BR /&gt;&lt;BR /&gt;Files:&lt;BR /&gt;        32453 created (91 per second)&lt;BR /&gt;                Creation alone: 30000 files (212 per second)&lt;BR /&gt;                Mixed with transactions: 2453 files (16 per second)&lt;BR /&gt;        2532 read (17 per second)&lt;BR /&gt;        2468 appended (16 per second)&lt;BR /&gt;        32453 deleted (91 per second)&lt;BR /&gt;                Deletion alone: 29906 files (439 per second)&lt;BR /&gt;                Mixed with transactions: 2547 files (17 per second)&lt;BR /&gt;&lt;BR /&gt;Data:&lt;BR /&gt;        12.67 megabytes read (36.54 kilobytes per second)&lt;BR /&gt;        167.31 megabytes written (482.60 kilobytes per second)&lt;BR /&gt;&lt;BR /&gt;Very Bad.</description>
      <pubDate>Fri, 05 Jul 2002 10:34:55 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757901#M5346</guid>
      <dc:creator>Domenico Viggiani</dc:creator>
      <dc:date>2002-07-05T10:34:55Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757902#M5347</link>
      <description>Hi&lt;BR /&gt;I agree, there must be something wrong, perhaps firmware upgrading will help.&lt;BR /&gt;&lt;BR /&gt;My test was from a CMD 7400 RAID controller and a RAID 5 set with only 4 drives. A VA7400 should perform much better.</description>
      <pubDate>Fri, 05 Jul 2002 11:08:54 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757902#M5347</guid>
      <dc:creator>Leif Halvarsson_2</dc:creator>
      <dc:date>2002-07-05T11:08:54Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757903#M5348</link>
      <description>These are results after upgrade to HP16:&lt;BR /&gt;&lt;BR /&gt;Time:&lt;BR /&gt;        164 seconds total&lt;BR /&gt;        66 seconds of transactions (75 per second)&lt;BR /&gt;&lt;BR /&gt;Files:&lt;BR /&gt;        32453 created (197 per second)&lt;BR /&gt;                Creation alone: 30000 files (535 per second)&lt;BR /&gt;                Mixed with transactions: 2453 files (37 per second)&lt;BR /&gt;        2532 read (38 per second)&lt;BR /&gt;        2468 appended (37 per second)&lt;BR /&gt;        32453 deleted (197 per second)&lt;BR /&gt;                Deletion alone: 29906 files (712 per second)&lt;BR /&gt;                Mixed with transactions: 2547 files (38 per second)&lt;BR /&gt;&lt;BR /&gt;Data:&lt;BR /&gt;        12.67 megabytes read (79.09 kilobytes per second)&lt;BR /&gt;        167.31 megabytes written (1.02 megabytes per second)</description>
      <pubDate>Wed, 17 Jul 2002 08:10:15 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757903#M5348</guid>
      <dc:creator>Domenico Viggiani</dc:creator>
      <dc:date>2002-07-17T08:10:15Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757904#M5349</link>
      <description>Hi Domenico,&lt;BR /&gt;I've seen some of your notes about troubles with VA7400. We 're evalueting of buying this storage system: would you mind having a private talk about its availability and performance? My address is pgenta@ferraniait.com &lt;BR /&gt;&lt;BR /&gt;Thanks&lt;BR /&gt;Paola Genta&lt;BR /&gt;</description>
      <pubDate>Thu, 18 Jul 2002 10:50:31 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757904#M5349</guid>
      <dc:creator>Genta Paola</dc:creator>
      <dc:date>2002-07-18T10:50:31Z</dc:date>
    </item>
    <item>
      <title>Re: VA7400 performance</title>
      <link>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757905#M5350</link>
      <description>We are using firmware revision HP15, and enclosure revision HP01.  We went thru firmware issues, and this one has been running fine for us in production for almost the past 2 months.&lt;BR /&gt;As for the oracle, there was an earlier post about breaking that part up.  I would have to agree.  Depending on the size of your database, you may see better performance by busting up your datafiles, and/or moveing your redo logs on to a separate lun.&lt;BR /&gt;</description>
      <pubDate>Thu, 18 Jul 2002 17:12:33 GMT</pubDate>
      <guid>https://community.hpe.com/t5/disk-enclosures/va7400-performance/m-p/2757905#M5350</guid>
      <dc:creator>Mark Malko</dc:creator>
      <dc:date>2002-07-18T17:12:33Z</dc:date>
    </item>
  </channel>
</rss>

