<?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 Slow EVA performance allocation unit/IOPS in HPE EVA Storage</title>
    <link>https://community.hpe.com/t5/hpe-eva-storage/slow-eva-performance-allocation-unit-iops/m-p/4638081#M44632</link>
    <description>We are getting high latencies and on our EVA8100.We have 56 1TB drives grouped, with vdisks being presented at 2TB.  2003 Server x86.  We tend to hit a read requests of 450-800+ Read Latency of 11-30+, AV Queue Depth of 10-20. I’m under the impression if the allocation unit size is changed from 4kb to 16kb+ on our backup server there would be much less IOPS. Most files on our backup servers are in the Gigabytes in size, there should be almost no loss in disk space. On my email servers I’m debating changing my allocation size from 4kb to 16kb for the same reason. Any advise will be appreciated.</description>
    <pubDate>Wed, 26 May 2010 15:28:50 GMT</pubDate>
    <dc:creator>Little Fish</dc:creator>
    <dc:date>2010-05-26T15:28:50Z</dc:date>
    <item>
      <title>Slow EVA performance allocation unit/IOPS</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/slow-eva-performance-allocation-unit-iops/m-p/4638081#M44632</link>
      <description>We are getting high latencies and on our EVA8100.We have 56 1TB drives grouped, with vdisks being presented at 2TB.  2003 Server x86.  We tend to hit a read requests of 450-800+ Read Latency of 11-30+, AV Queue Depth of 10-20. I’m under the impression if the allocation unit size is changed from 4kb to 16kb+ on our backup server there would be much less IOPS. Most files on our backup servers are in the Gigabytes in size, there should be almost no loss in disk space. On my email servers I’m debating changing my allocation size from 4kb to 16kb for the same reason. Any advise will be appreciated.</description>
      <pubDate>Wed, 26 May 2010 15:28:50 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/slow-eva-performance-allocation-unit-iops/m-p/4638081#M44632</guid>
      <dc:creator>Little Fish</dc:creator>
      <dc:date>2010-05-26T15:28:50Z</dc:date>
    </item>
    <item>
      <title>Re: Slow EVA performance allocation unit/IOPS</title>
      <link>https://community.hpe.com/t5/hpe-eva-storage/slow-eva-performance-allocation-unit-iops/m-p/4638082#M44633</link>
      <description>That math I came up with was 66fata disks * 34-IOPS at vRAID 5 = 2244-IOPS. 2244-IOPS * 4KB Clusters = 8976 KB.  When I check the logs the SAN throughput peeks 5MB +/- 93MB.  If I increase the cluster size to 16KB 35904KB throughput @ 32KB cluster throughput 71808KB. Lost of disk space would be negligable.</description>
      <pubDate>Wed, 26 May 2010 17:44:47 GMT</pubDate>
      <guid>https://community.hpe.com/t5/hpe-eva-storage/slow-eva-performance-allocation-unit-iops/m-p/4638082#M44633</guid>
      <dc:creator>Little Fish</dc:creator>
      <dc:date>2010-05-26T17:44:47Z</dc:date>
    </item>
  </channel>
</rss>

