<?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: VLS9000 write performance in StoreOnce Backup Storage</title>
    <link>https://community.hpe.com/t5/storeonce-backup-storage/vls9000-write-performance/m-p/4558849#M957</link>
    <description>I think you are confusing the D2D Arrays with the VLS. Data Dedup is not done during transfer on the VLS, it is done after a full cartridge is written to so it does not affect performance of writes at all. You can find the specs at &lt;A href="http://h18000.www1.hp.com/products/quickspecs/12812_div/12812_div.HTML." target="_blank"&gt;http://h18000.www1.hp.com/products/quickspecs/12812_div/12812_div.HTML.&lt;/A&gt; &lt;BR /&gt;&lt;BR /&gt;I have found by adding more (virtual) tape drives to the (virtual) ESL Library and forcing more streams from the backup client has improved performance so far to almost 300MB/s. I am still doing some testing so I will post back with the results if anyone is interested.</description>
    <pubDate>Wed, 13 Jan 2010 17:37:05 GMT</pubDate>
    <dc:creator>ddaloia</dc:creator>
    <dc:date>2010-01-13T17:37:05Z</dc:date>
    <item>
      <title>VLS9000 write performance</title>
      <link>https://community.hpe.com/t5/storeonce-backup-storage/vls9000-write-performance/m-p/4558847#M955</link>
      <description>&lt;P&gt;Hi Folks,&lt;BR /&gt;&lt;BR /&gt;Just getting started with the VLS9000 and we notice any backups we write to tape max out at 100MB/s. This happens no matter how many servers, jobs, or tapes we are writing to at one time. It is a single node system which HP advertises as capable of 550MB/s with no compression or 600MB/s with 2:1 compression. &lt;BR /&gt;&lt;BR /&gt;My question is what are other folks getting for performance and what are some things I can do to get better performance?&lt;BR /&gt;&lt;BR /&gt;Just to add, the backup storage node is a BL460G6, 2-4core 5570 procs, 6GB mem, dual 10GbE nics. All other servers we are backing up are similar, only way more memory. Tested backups with Data Protector and EMC Networker.&lt;BR /&gt;&lt;BR /&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;P.S. This thread has been moevd&amp;nbsp;from Tape Libraries and Drives to Disk-Based Backup. - Hp Forum moderator&lt;/P&gt;</description>
      <pubDate>Mon, 04 Aug 2014 05:01:55 GMT</pubDate>
      <guid>https://community.hpe.com/t5/storeonce-backup-storage/vls9000-write-performance/m-p/4558847#M955</guid>
      <dc:creator>ddaloia</dc:creator>
      <dc:date>2014-08-04T05:01:55Z</dc:date>
    </item>
    <item>
      <title>Re: VLS9000 write performance</title>
      <link>https://community.hpe.com/t5/storeonce-backup-storage/vls9000-write-performance/m-p/4558848#M956</link>
      <description>Hi,&lt;BR /&gt;where did you got these figure?&lt;BR /&gt;I thought the max  speed with only one single array without dedup was max 240 MB/sec &lt;BR /&gt;and with dedup (FW 3.x.x) max 150 MB/s&lt;BR /&gt;You should considere also limitation due to the source of the data, the SAN path, etc...</description>
      <pubDate>Wed, 13 Jan 2010 16:13:29 GMT</pubDate>
      <guid>https://community.hpe.com/t5/storeonce-backup-storage/vls9000-write-performance/m-p/4558848#M956</guid>
      <dc:creator>Marino Meloni_1</dc:creator>
      <dc:date>2010-01-13T16:13:29Z</dc:date>
    </item>
    <item>
      <title>Re: VLS9000 write performance</title>
      <link>https://community.hpe.com/t5/storeonce-backup-storage/vls9000-write-performance/m-p/4558849#M957</link>
      <description>I think you are confusing the D2D Arrays with the VLS. Data Dedup is not done during transfer on the VLS, it is done after a full cartridge is written to so it does not affect performance of writes at all. You can find the specs at &lt;A href="http://h18000.www1.hp.com/products/quickspecs/12812_div/12812_div.HTML." target="_blank"&gt;http://h18000.www1.hp.com/products/quickspecs/12812_div/12812_div.HTML.&lt;/A&gt; &lt;BR /&gt;&lt;BR /&gt;I have found by adding more (virtual) tape drives to the (virtual) ESL Library and forcing more streams from the backup client has improved performance so far to almost 300MB/s. I am still doing some testing so I will post back with the results if anyone is interested.</description>
      <pubDate>Wed, 13 Jan 2010 17:37:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/storeonce-backup-storage/vls9000-write-performance/m-p/4558849#M957</guid>
      <dc:creator>ddaloia</dc:creator>
      <dc:date>2010-01-13T17:37:05Z</dc:date>
    </item>
    <item>
      <title>Re: VLS9000 write performance</title>
      <link>https://community.hpe.com/t5/storeonce-backup-storage/vls9000-write-performance/m-p/4558850#M958</link>
      <description>Just as an FYI, I've had a fair amount of experience with dedup on the 6000's (I'm upgrading to a 9000 today) and it's full tapes that get deduped. To reduce this I usually use one drive at a time for a backup to enable the best dedup performance. As I grow our backups, I can see that writing to more then one tape at a time may not really hurt the deduping. But if you were writing to 2 tapes at once you'll need to fill both tapes before it'll dedup that backup at all. For a large backup this may not be bad especially if you filled it up daily. But otherwise it could be a problem.</description>
      <pubDate>Wed, 12 May 2010 17:34:54 GMT</pubDate>
      <guid>https://community.hpe.com/t5/storeonce-backup-storage/vls9000-write-performance/m-p/4558850#M958</guid>
      <dc:creator>Kyle Weir</dc:creator>
      <dc:date>2010-05-12T17:34:54Z</dc:date>
    </item>
  </channel>
</rss>

