<?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 Disk performance testing in HPE EVA Storage</title>
    <link>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918845#M2092</link>
    <description>I want to do performance testing from my server to disks residing on a EMC SAN storage system.  In the past I have used the timex dd command but not sure if this is an accurate test to the EMC box.  I am using META volumes and have broken my spendels into Super High, High, Medium and low.&lt;BR /&gt;&lt;BR /&gt;What command or commands can I use to get an accurate performace test?</description>
    <pubDate>Tue, 04 Mar 2003 18:57:37 GMT</pubDate>
    <dc:creator>Rick Barr_1</dc:creator>
    <dc:date>2003-03-04T18:57:37Z</dc:date>
    <item>
      <title>Disk performance testing</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918845#M2092</link>
      <description>I want to do performance testing from my server to disks residing on a EMC SAN storage system.  In the past I have used the timex dd command but not sure if this is an accurate test to the EMC box.  I am using META volumes and have broken my spendels into Super High, High, Medium and low.&lt;BR /&gt;&lt;BR /&gt;What command or commands can I use to get an accurate performace test?</description>
      <pubDate>Tue, 04 Mar 2003 18:57:37 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918845#M2092</guid>
      <dc:creator>Rick Barr_1</dc:creator>
      <dc:date>2003-03-04T18:57:37Z</dc:date>
    </item>
    <item>
      <title>Re: Disk performance testing</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918846#M2093</link>
      <description>Lots of commands will do it.&lt;BR /&gt;&lt;BR /&gt;I sometimes issue a series of complex find commands starting at the root.  this load tests and gets a not of processes hitting the disk.&lt;BR /&gt;&lt;BR /&gt;If there are database apps on the disk, I reindex keys at varying levels based on how hard I want to push the disk.&lt;BR /&gt;&lt;BR /&gt;I collect the data with this attached script.&lt;BR /&gt;&lt;BR /&gt;SEP</description>
      <pubDate>Tue, 04 Mar 2003 19:01:27 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918846#M2093</guid>
      <dc:creator>Steven E. Protter</dc:creator>
      <dc:date>2003-03-04T19:01:27Z</dc:date>
    </item>
    <item>
      <title>Re: Disk performance testing</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918847#M2094</link>
      <description>Hi Rick,&lt;BR /&gt;&lt;BR /&gt;One thing I have observed so far is that any load test you do at the ground level is not going to be too valid once you load the application. The performance is dependent on how the application accesses the data and how you laid out your filesystems.&lt;BR /&gt;&lt;BR /&gt;I would suggest to load your actual application and run load tests through it. Try with various scenarios like online JFS options, seperation of filesystems per  activity etc., Choose the ones that give you the best performance for "your application".&lt;BR /&gt;&lt;BR /&gt;-Sri</description>
      <pubDate>Tue, 04 Mar 2003 19:20:14 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918847#M2094</guid>
      <dc:creator>Sridhar Bhaskarla</dc:creator>
      <dc:date>2003-03-04T19:20:14Z</dc:date>
    </item>
    <item>
      <title>Re: Disk performance testing</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918848#M2095</link>
      <description>I've attached a command, pat, that I creatively acquired from a HP tuning guy.  The command "pat -f /mount/point" gives you some great information.  Its as dumb as a rock, but I like it: it gives simple numbers that even pointy-haired bosses can sometimes understand--but don't count on it. &lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Chris</description>
      <pubDate>Tue, 04 Mar 2003 20:20:11 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918848#M2095</guid>
      <dc:creator>Chris Vail</dc:creator>
      <dc:date>2003-03-04T20:20:11Z</dc:date>
    </item>
    <item>
      <title>Re: Disk performance testing</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918849#M2096</link>
      <description>This is the tool which can give you great results&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://www.storagetek.com/prodserv/products/software/freeware/stkio/" target="_blank"&gt;http://www.storagetek.com/prodserv/products/software/freeware/stkio/&lt;/A&gt;</description>
      <pubDate>Wed, 05 Mar 2003 05:49:12 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918849#M2096</guid>
      <dc:creator>T G Manikandan</dc:creator>
      <dc:date>2003-03-05T05:49:12Z</dc:date>
    </item>
    <item>
      <title>Re: Disk performance testing</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918850#M2097</link>
      <description>Hi,&lt;BR /&gt;We've just had extensive consultancy on this issue (with consultants from both HP and EMC onsite). You've got to distinguish between what you want to test.&lt;BR /&gt;&lt;BR /&gt;1. Throughput&lt;BR /&gt;&lt;A href="http://www.iozone.org/" target="_blank"&gt;http://www.iozone.org/&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;2. Response time &lt;BR /&gt;&lt;A href="ftp://ftp.emc.com/pub/symm3000/iorate/" target="_blank"&gt;ftp://ftp.emc.com/pub/symm3000/iorate/&lt;/A&gt;&lt;BR /&gt;# Read the manpage very carefully, this tool is potentially destructive !&lt;BR /&gt;&lt;BR /&gt;If you want very specific data, you might look at EMC's Workload Analyzer (WLA). However, you've got to pay to get that ...&lt;BR /&gt;&lt;BR /&gt;The HP guys based their interpretation mostly on data from sar (-d). The avserv and avwait from sar can also be used to compute average response times (just add them ;-) in the interval.&lt;BR /&gt;&lt;BR /&gt;This should put you "on the road".&lt;BR /&gt;Regards,&lt;BR /&gt;Tom</description>
      <pubDate>Wed, 05 Mar 2003 06:46:52 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918850#M2097</guid>
      <dc:creator>Tom Geudens</dc:creator>
      <dc:date>2003-03-05T06:46:52Z</dc:date>
    </item>
    <item>
      <title>Re: Disk performance testing</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918851#M2098</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;i think you are looking for end to end performence testing of your environment, infact you have to use command like dd or volume manager related command like pvmove or veristas commands like relayout or so.&lt;BR /&gt;but if you want to do the testing on EMC disk/Hyper performencs you WLA (Work load analyzer or Symtop ig good tools from EMC. apart of this if you have SYMCLI installed you can use symstat command it will give you i/o operations on each luns.&lt;BR /&gt;&lt;BR /&gt;Regards&lt;BR /&gt;&lt;BR /&gt;Sunil</description>
      <pubDate>Fri, 21 Mar 2003 05:32:17 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918851#M2098</guid>
      <dc:creator>Sunil Sharma_1</dc:creator>
      <dc:date>2003-03-21T05:32:17Z</dc:date>
    </item>
    <item>
      <title>Re: Disk performance testing</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918852#M2099</link>
      <description>Please do not use dd for perfomance testing. dd is single threaded, and does not create concurrent I/Os. Of course, you can run multiple instances of dd, but in this case perfomance may be difficult to calculate. I would recommend to use specialized perfomance measurement tools, not these dumb like dd in unix and copy/xcopy in windows&lt;BR /&gt;Eugeny</description>
      <pubDate>Fri, 21 Mar 2003 07:09:44 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/disk-performance-testing/m-p/2918852#M2099</guid>
      <dc:creator>Eugeny Brychkov</dc:creator>
      <dc:date>2003-03-21T07:09:44Z</dc:date>
    </item>
  </channel>
</rss>

