<?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: performance problem using host-based-mirroring over two computer centers in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-using-host-based-mirroring-over-two-computer/m-p/2757611#M643365</link>
    <description>Hi Roland,&lt;BR /&gt;&lt;BR /&gt;First place I'd look is your SAN config:&lt;BR /&gt;&lt;BR /&gt;1. Have you checked that all devices are in fabric login mode, and that nothing is using FC-AL (loop topology). FCAL provides 100Mb/s bandwidth shared between all devices. Fabric login, or fabric mode provides 100Mb/s to each device.&lt;BR /&gt;&lt;BR /&gt;2. Your two ISL links - what type of links are these? Longwave or shortwave GBICs? What type of fibre provides the connections? 9/125, 50/125 or 62.5/125? Or are you using DWDM?&lt;BR /&gt;&lt;BR /&gt;3. Have you checked for execssive error logging on the switches? -Sorry I don't have my brocade notes with me so I can't remember the commands to gather this info...&lt;BR /&gt;&lt;BR /&gt;Next we should probably take a look at your logical volume configuration - can you post a vgdisplay for one of your mirrored volume groups, and a lvdisplay for one of the volumes in that volume group.&lt;BR /&gt;&lt;BR /&gt;That should at least give us some food for thought...&lt;BR /&gt;&lt;BR /&gt;Duncan</description>
    <pubDate>Thu, 04 Jul 2002 07:03:21 GMT</pubDate>
    <dc:creator>Duncan Edmonstone</dc:creator>
    <dc:date>2002-07-04T07:03:21Z</dc:date>
    <item>
      <title>performance problem using host-based-mirroring over two computer centers</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-using-host-based-mirroring-over-two-computer/m-p/2757610#M643364</link>
      <description>Hi admins :-)&lt;BR /&gt;&lt;BR /&gt;We have some performance problem using host-based-mirroring over two computer centers.&lt;BR /&gt;&lt;BR /&gt;We have two data centers, in each there is a L-Class HPUX11.i, two Brocade FC-switches and one VA7100. Between the switches there are two InterSwitchLinks, so each server can also 'see' the storage in the other data center. Using LVM host-based-mirroring for an shared oracle volumegroup (rawdevices) we got a two cell ServiceGuard-Cluster with Oracle-Parallel-Server functionality.&lt;BR /&gt;&lt;BR /&gt;But now we discovered a slow io rate during LVM-mirroring. A simple file creating using dd-command needs much more time (times 4) on mirrored disks than on the local disks ! And it is hard to discuss this with angry application people. So my question to you : do you have any suggestions or ideas where to look ? Which parameters to change ? How to increase performance ? How to get more detailed infos where exactly we got the problem ? What to change in hardware ?&lt;BR /&gt;&lt;BR /&gt;Any information is welcome ! Thanks for your answers to my mail, i will summarize as soon as possible !&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;An admin in trouble&lt;BR /&gt;Roland&lt;BR /&gt;:-)&lt;BR /&gt;</description>
      <pubDate>Thu, 04 Jul 2002 03:49:49 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-using-host-based-mirroring-over-two-computer/m-p/2757610#M643364</guid>
      <dc:creator>OSY Unix 1-4-2 (A-Team)</dc:creator>
      <dc:date>2002-07-04T03:49:49Z</dc:date>
    </item>
    <item>
      <title>Re: performance problem using host-based-mirroring over two computer centers</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-using-host-based-mirroring-over-two-computer/m-p/2757611#M643365</link>
      <description>Hi Roland,&lt;BR /&gt;&lt;BR /&gt;First place I'd look is your SAN config:&lt;BR /&gt;&lt;BR /&gt;1. Have you checked that all devices are in fabric login mode, and that nothing is using FC-AL (loop topology). FCAL provides 100Mb/s bandwidth shared between all devices. Fabric login, or fabric mode provides 100Mb/s to each device.&lt;BR /&gt;&lt;BR /&gt;2. Your two ISL links - what type of links are these? Longwave or shortwave GBICs? What type of fibre provides the connections? 9/125, 50/125 or 62.5/125? Or are you using DWDM?&lt;BR /&gt;&lt;BR /&gt;3. Have you checked for execssive error logging on the switches? -Sorry I don't have my brocade notes with me so I can't remember the commands to gather this info...&lt;BR /&gt;&lt;BR /&gt;Next we should probably take a look at your logical volume configuration - can you post a vgdisplay for one of your mirrored volume groups, and a lvdisplay for one of the volumes in that volume group.&lt;BR /&gt;&lt;BR /&gt;That should at least give us some food for thought...&lt;BR /&gt;&lt;BR /&gt;Duncan</description>
      <pubDate>Thu, 04 Jul 2002 07:03:21 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-using-host-based-mirroring-over-two-computer/m-p/2757611#M643365</guid>
      <dc:creator>Duncan Edmonstone</dc:creator>
      <dc:date>2002-07-04T07:03:21Z</dc:date>
    </item>
    <item>
      <title>Re: performance problem using host-based-mirroring over two computer centers</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-using-host-based-mirroring-over-two-computer/m-p/2757612#M643366</link>
      <description>Also Roland you should consider posting this query in the Storage forum here:&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums.itrc.hp.com/cm/FamilyHome/1,,120,00.html" target="_blank"&gt;http://forums.itrc.hp.com/cm/FamilyHome/1,,120,00.html&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;Not all the experts there necessarily read the HP-UX forum posts...&lt;BR /&gt;&lt;BR /&gt;HTH&lt;BR /&gt;&lt;BR /&gt;Duncan</description>
      <pubDate>Thu, 04 Jul 2002 07:10:55 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-problem-using-host-based-mirroring-over-two-computer/m-p/2757612#M643366</guid>
      <dc:creator>Duncan Edmonstone</dc:creator>
      <dc:date>2002-07-04T07:10:55Z</dc:date>
    </item>
  </channel>
</rss>

