<?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 Performance Through Disk Striping in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-through-disk-striping/m-p/4150178#M599604</link>
    <description>Hi All;&lt;BR /&gt;&lt;BR /&gt;I have done striping on 12 x 73 GB on external HPstorage boxe which is connected to the server server through two cards fiber;&lt;BR /&gt;So I waited that the  data written would be very fast on the logical volume, unfortunately the rate of write data is very slow on the logical volume .&lt;BR /&gt;I give an example : for writing 632GB of data it tooks about 15 hours to finish; can you tell me what is the wrong in my configuration :&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;--- Volume groups ---&lt;BR /&gt;VG Name                     /dev/vg08&lt;BR /&gt;VG Write Access             read/write&lt;BR /&gt;VG Status                   available&lt;BR /&gt;Max LV                      255&lt;BR /&gt;Cur LV                      1&lt;BR /&gt;Open LV                     1&lt;BR /&gt;Max PV                      16&lt;BR /&gt;Cur PV                      12&lt;BR /&gt;Act PV                      12&lt;BR /&gt;Max PE per PV               65000&lt;BR /&gt;VGDA                        24&lt;BR /&gt;PE Size (Mbytes)            16&lt;BR /&gt;Total PE                    52488&lt;BR /&gt;Alloc PE                    52488&lt;BR /&gt;Free PE                     0&lt;BR /&gt;Total PVG                   0&lt;BR /&gt;Total Spare PVs             0&lt;BR /&gt;Total Spare PVs in use      0&lt;BR /&gt;&lt;BR /&gt;--- Logical volumes ---&lt;BR /&gt;LV Name                     /dev/vg08/lvol1&lt;BR /&gt;VG Name                     /dev/vg08&lt;BR /&gt;LV Permission               read/write&lt;BR /&gt;LV Status                   available/syncd&lt;BR /&gt;Mirror copies               0&lt;BR /&gt;Consistency Recovery        MWC&lt;BR /&gt;Schedule                    striped&lt;BR /&gt;LV Size (Mbytes)            839808&lt;BR /&gt;Current LE                  52488&lt;BR /&gt;Allocated PE                52488&lt;BR /&gt;Stripes                     12&lt;BR /&gt;Stripe Size (Kbytes)        64&lt;BR /&gt;Bad block                   on&lt;BR /&gt;Allocation                  strict&lt;BR /&gt;IO Timeout (Seconds)        default&lt;BR /&gt;&lt;BR /&gt;   --- Distribution of logical volume ---&lt;BR /&gt;   PV Name            LE on PV  PE on PV&lt;BR /&gt;   /dev/dsk/c18t1d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t2d0   4374      4374&lt;BR /&gt;   /dev/dsk/c18t3d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t4d0   4374      4374&lt;BR /&gt;   /dev/dsk/c18t5d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t6d0   4374      4374&lt;BR /&gt;   /dev/dsk/c18t7d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t8d0   4374      4374&lt;BR /&gt;   /dev/dsk/c18t9d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t10d0  4374      4374&lt;BR /&gt;   /dev/dsk/c18t11d0  4374      4374&lt;BR /&gt;   /dev/dsk/c19t12d0  4374      4374&lt;BR /&gt;&lt;BR /&gt;many thanks.&lt;BR /&gt;</description>
    <pubDate>Mon, 25 Feb 2008 18:40:06 GMT</pubDate>
    <dc:creator>AntriBouzar</dc:creator>
    <dc:date>2008-02-25T18:40:06Z</dc:date>
    <item>
      <title>Performance Through Disk Striping</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-through-disk-striping/m-p/4150178#M599604</link>
      <description>Hi All;&lt;BR /&gt;&lt;BR /&gt;I have done striping on 12 x 73 GB on external HPstorage boxe which is connected to the server server through two cards fiber;&lt;BR /&gt;So I waited that the  data written would be very fast on the logical volume, unfortunately the rate of write data is very slow on the logical volume .&lt;BR /&gt;I give an example : for writing 632GB of data it tooks about 15 hours to finish; can you tell me what is the wrong in my configuration :&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;--- Volume groups ---&lt;BR /&gt;VG Name                     /dev/vg08&lt;BR /&gt;VG Write Access             read/write&lt;BR /&gt;VG Status                   available&lt;BR /&gt;Max LV                      255&lt;BR /&gt;Cur LV                      1&lt;BR /&gt;Open LV                     1&lt;BR /&gt;Max PV                      16&lt;BR /&gt;Cur PV                      12&lt;BR /&gt;Act PV                      12&lt;BR /&gt;Max PE per PV               65000&lt;BR /&gt;VGDA                        24&lt;BR /&gt;PE Size (Mbytes)            16&lt;BR /&gt;Total PE                    52488&lt;BR /&gt;Alloc PE                    52488&lt;BR /&gt;Free PE                     0&lt;BR /&gt;Total PVG                   0&lt;BR /&gt;Total Spare PVs             0&lt;BR /&gt;Total Spare PVs in use      0&lt;BR /&gt;&lt;BR /&gt;--- Logical volumes ---&lt;BR /&gt;LV Name                     /dev/vg08/lvol1&lt;BR /&gt;VG Name                     /dev/vg08&lt;BR /&gt;LV Permission               read/write&lt;BR /&gt;LV Status                   available/syncd&lt;BR /&gt;Mirror copies               0&lt;BR /&gt;Consistency Recovery        MWC&lt;BR /&gt;Schedule                    striped&lt;BR /&gt;LV Size (Mbytes)            839808&lt;BR /&gt;Current LE                  52488&lt;BR /&gt;Allocated PE                52488&lt;BR /&gt;Stripes                     12&lt;BR /&gt;Stripe Size (Kbytes)        64&lt;BR /&gt;Bad block                   on&lt;BR /&gt;Allocation                  strict&lt;BR /&gt;IO Timeout (Seconds)        default&lt;BR /&gt;&lt;BR /&gt;   --- Distribution of logical volume ---&lt;BR /&gt;   PV Name            LE on PV  PE on PV&lt;BR /&gt;   /dev/dsk/c18t1d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t2d0   4374      4374&lt;BR /&gt;   /dev/dsk/c18t3d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t4d0   4374      4374&lt;BR /&gt;   /dev/dsk/c18t5d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t6d0   4374      4374&lt;BR /&gt;   /dev/dsk/c18t7d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t8d0   4374      4374&lt;BR /&gt;   /dev/dsk/c18t9d0   4374      4374&lt;BR /&gt;   /dev/dsk/c19t10d0  4374      4374&lt;BR /&gt;   /dev/dsk/c18t11d0  4374      4374&lt;BR /&gt;   /dev/dsk/c19t12d0  4374      4374&lt;BR /&gt;&lt;BR /&gt;many thanks.&lt;BR /&gt;</description>
      <pubDate>Mon, 25 Feb 2008 18:40:06 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-through-disk-striping/m-p/4150178#M599604</guid>
      <dc:creator>AntriBouzar</dc:creator>
      <dc:date>2008-02-25T18:40:06Z</dc:date>
    </item>
    <item>
      <title>Re: Performance Through Disk Striping</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-through-disk-striping/m-p/4150179#M599605</link>
      <description>Without getting more details I would say that you have an I/O bus issue right now probably because of overstriping. Striping spreads the I/O on multiple physical disks to take advantage of the I/O bandwidth of the added spindles. However all these spindles are sharing the same I/O bus (the same two I/O buses in your case) There is probably a lot of queueing happening in your server in trying to update all these spindles at once.&lt;BR /&gt;&lt;BR /&gt;You should experiment with using fewer spindles in your striping. &lt;BR /&gt;&lt;BR /&gt;Also did you use extend striping or regular (aka KB striping, aka "-i -I") striping. &lt;BR /&gt;&lt;BR /&gt;I would say that even with 15kRPM disks you should not use more than 4 striped spindles in your LVs (2 spindles per controller). &lt;BR /&gt;&lt;BR /&gt;If you have the flexibility, experiment with different number of spindles.</description>
      <pubDate>Mon, 25 Feb 2008 19:30:39 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-through-disk-striping/m-p/4150179#M599605</guid>
      <dc:creator>TTr</dc:creator>
      <dc:date>2008-02-25T19:30:39Z</dc:date>
    </item>
  </channel>
</rss>

