<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">On 05/05/2017 08:42 PM, Pat Haley
      wrote:<br>
    </div>
    <blockquote cite="mid:b1f4209e-a0e5-5df4-f954-399df8270b21@mit.edu"
      type="cite">
      <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
      <br>
      Hi Pranith,<br>
      <br>
      I presume you are asking for some version of the profile data that
      just shows the dd test (or a repeat of the dd test).  If yes, how
      do I extract just that data?<br>
    </blockquote>
    Yes, that is what he is asking for. Just clear the existing profile
    info using `gluster volume profile volname clear` and run the dd
    test once. Then when you run profile info again, it should just give
    you the stats for the dd test.<br>
    <blockquote cite="mid:b1f4209e-a0e5-5df4-f954-399df8270b21@mit.edu"
      type="cite"> <br>
      Thanks<br>
      <br>
      Pat<br>
      <br>
      <br>
      <br>
      <div class="moz-cite-prefix">On 05/05/2017 10:58 AM, Pranith Kumar
        Karampuri wrote:<br>
      </div>
      <blockquote
cite="mid:CAOgeEnYKv+v=Dv3y0tC9WVJSfN=dH2U9QmrXqwOOvbTGZD=yWw@mail.gmail.com"
        type="cite">
        <meta http-equiv="Content-Type" content="text/html;
          charset=utf-8">
        <div dir="ltr">
          <div>
            <div>
              <div>hi Pat,<br>
              </div>
                    Let us concentrate on the performance numbers part
              for now. We will look at the permissions one after this?<br>
              <br>
            </div>
            As per the profile info, only 2.6% of the work-load is
            writes. There are too many Lookups.<br>
            <br>
          </div>
          Would it be possible to get the data for just the dd test you
          were doing earlier?<br>
          <br>
        </div>
        <div class="gmail_extra"><br>
          <div class="gmail_quote">On Fri, May 5, 2017 at 8:14 PM, Pat
            Haley <span dir="ltr">&lt;<a moz-do-not-send="true"
                href="mailto:phaley@mit.edu" target="_blank">phaley@mit.edu</a>&gt;</span>
            wrote:<br>
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div bgcolor="#FFFFFF" text="#000000"> <br>
                Hi Pranith &amp; Ravi,<br>
                <br>
                A couple of quick questions<br>
                <br>
                We have profile turned on. Are there specific queries we
                should make that would help debug our configuration? 
                (The default profile info was previously sent in <a
                  moz-do-not-send="true"
                  class="m_-8630651799581052987moz-txt-link-freetext"
href="http://lists.gluster.org/pipermail/gluster-users/2017-May/030840.html"
                  target="_blank">http://lists.gluster.org/<wbr>pipermail/gluster-users/2017-<wbr>May/030840.html</a>
                but I'm not sure if that is what you were looking for.)<br>
                <br>
                We also started to do a test on serving gluster over
                NFS.  We rediscovered an issue we previously reported (
                <a moz-do-not-send="true"
                  class="m_-8630651799581052987moz-txt-link-freetext"
href="http://lists.gluster.org/pipermail/gluster-users/2016-September/028289.html"
                  target="_blank">http://lists.gluster.org/<wbr>pipermail/gluster-users/2016-<wbr>September/028289.html</a>
                ) in that the NFS mounted version was ignoring the group
                write permissions.  What specific information would be
                useful in debugging this?<br>
                <br>
                Thanks<span class="HOEnZb"><font color="#888888"><br>
                    <br>
                    Pat</font></span>
                <div>
                  <div class="h5"><br>
                    <br>
                    <br>
                    <div class="m_-8630651799581052987moz-cite-prefix">On
                      04/14/2017 03:01 AM, Ravishankar N wrote:<br>
                    </div>
                    <blockquote type="cite">
                      <div class="m_-8630651799581052987moz-cite-prefix">On
                        04/14/2017 12:20 PM, Pranith Kumar Karampuri
                        wrote:<br>
                      </div>
                      <blockquote type="cite">
                        <div dir="ltr"><br>
                          <div class="gmail_extra"><br>
                            <div class="gmail_quote">On Sat, Apr 8, 2017
                              at 10:28 AM, Ravishankar N <span
                                dir="ltr">&lt;<a moz-do-not-send="true"
                                  href="mailto:ravishankar@redhat.com"
                                  target="_blank">ravishankar@redhat.com</a>&gt;</span>
                              wrote:<br>
                              <blockquote class="gmail_quote"
                                style="margin:0px 0px 0px
                                0.8ex;border-left:1px solid
                                rgb(204,204,204);padding-left:1ex">
                                <div bgcolor="#FFFFFF">
                                  <div
                                    class="m_-8630651799581052987gmail-m_1278894059907384689moz-cite-prefix">Hi
                                    Pat,<br>
                                    <br>
                                    I'm assuming you are using gluster
                                    native (fuse mount). If it helps,
                                    you could try mounting it via
                                    gluster NFS (gnfs) and then see if
                                    there is an improvement in speed.
                                    Fuse mounts are slower than gnfs
                                    mounts but you get the benefit of
                                    avoiding a single point of failure.
                                    Unlike fuse mounts, if the gluster
                                    node containing the gnfs server goes
                                    down, all mounts done using that
                                    node will fail). For fuse mounts,
                                    you could try tweaking the
                                    write-behind xlator settings to see
                                    if it helps. See the
                                    performance.write-behind and
                                    performance.write-behind-windo<wbr>w-size
                                    options in `gluster volume set
                                    help`. Of course, even for gnfs
                                    mounts, you can achieve fail-over by
                                    using CTDB.<br>
                                  </div>
                                </div>
                              </blockquote>
                              <div><br>
                              </div>
                              <div>Ravi,<br>
                              </div>
                              <div>      Do you have any data that
                                suggests fuse mounts are slower than
                                gNFS servers? <br>
                              </div>
                            </div>
                          </div>
                        </div>
                      </blockquote>
                      I have heard anecdotal evidence time and again on
                      the ML and IRC, which is why I wanted to compare
                      it with NFS numbers on his setup. <br>
                      <blockquote type="cite">
                        <div dir="ltr">
                          <div class="gmail_extra">
                            <div class="gmail_quote">
                              <div><br>
                              </div>
                              <div>Pat,<br>
                              </div>
                              <div>      I see that I am late to the
                                thread, but do you happen to have
                                "profile info" of the workload?<br>
                                <br>
                              </div>
                              <div>You can follow <a
                                  moz-do-not-send="true"
href="https://gluster.readthedocs.io/en/latest/Administrator%20Guide/Monitoring%20Workload/"
                                  target="_blank">https://gluster.readthedocs.<wbr>io/en/latest/Administrator%<wbr>20Guide/Monitoring%20Workload/</a>
                                to get the information.<br>
                              </div>
                            </div>
                          </div>
                        </div>
                      </blockquote>
                      Yeah, Let's see if profile info shows up anything
                      interesting.<br>
                      -Ravi<br>
                      <blockquote type="cite">
                        <div dir="ltr">
                          <div class="gmail_extra">
                            <div class="gmail_quote">
                              <div> </div>
                              <blockquote class="gmail_quote"
                                style="margin:0px 0px 0px
                                0.8ex;border-left:1px solid
                                rgb(204,204,204);padding-left:1ex">
                                <div bgcolor="#FFFFFF">
                                  <div
                                    class="m_-8630651799581052987gmail-m_1278894059907384689moz-cite-prefix">
                                    <br>
                                    Thanks,<br>
                                    Ravi
                                    <div>
                                      <div
                                        class="m_-8630651799581052987gmail-h5"><br>
                                        <br>
                                        On 04/08/2017 12:07 AM, Pat
                                        Haley wrote:<br>
                                      </div>
                                    </div>
                                  </div>
                                  <blockquote type="cite">
                                    <div>
                                      <div
                                        class="m_-8630651799581052987gmail-h5">
                                        <br>
                                        Hi,<br>
                                        <br>
                                        We noticed a dramatic slowness
                                        when writing to a gluster disk
                                        when compared to writing to an
                                        NFS disk. Specifically when
                                        using dd (data duplicator) to
                                        write a 4.3 GB file of zeros:<br>
                                        <ul>
                                          <li>on NFS disk (/home): 9.5
                                            Gb/s</li>
                                          <li>on gluster disk (/gdata):
                                            508 Mb/s<br>
                                          </li>
                                        </ul>
                                        The gluser disk is 2 bricks
                                        joined together, no replication
                                        or anything else. The hardware
                                        is (literally) the same:<br>
                                        <ul>
                                          <li>one server with 70 hard
                                            disks  and a hardware RAID
                                            card.</li>
                                          <li>4 disks in a RAID-6 group
                                            (the NFS disk)</li>
                                          <li>32 disks in a RAID-6 group
                                            (the max allowed by the
                                            card, /mnt/brick1)</li>
                                          <li>32 disks in another RAID-6
                                            group (/mnt/brick2)</li>
                                          <li>2 hot spare<br>
                                          </li>
                                        </ul>
                                        <p>Some additional information
                                          and more tests results (after
                                          changing the log level):<br>
                                        </p>
                                        <p><span>glusterfs 3.7.11 built
                                            on Apr 27 2016 14:09:22</span><br>
                                          <span>CentOS release 6.8
                                            (Final)</span><br>
                                          RAID bus controller: LSI Logic
                                          / Symbios Logic MegaRAID SAS-3
                                          3108 [Invader] (rev 02)<br>
                                          <br>
                                          <br>
                                          <br>
                                          <b>Create the file to /gdata
                                            (gluster)</b><br>
                                          [root@mseas-data2 gdata]# dd
                                          if=/dev/zero of=/gdata/zero1
                                          bs=1M count=1000<br>
                                          1000+0 records in<br>
                                          1000+0 records out<br>
                                          1048576000 bytes (1.0 GB)
                                          copied, 1.91876 s, <b>546
                                            MB/s</b><br>
                                          <br>
                                          <b>Create the file to /home
                                            (ext4)</b><br>
                                          [root@mseas-data2 gdata]# dd
                                          if=/dev/zero of=/home/zero1
                                          bs=1M count=1000<br>
                                          1000+0 records in<br>
                                          1000+0 records out<br>
                                          1048576000 bytes (1.0 GB)
                                          copied, 0.686021 s, <b>1.5
                                            GB/s - </b>3 times as fast<b><br>
                                            <br>
                                            <br>
                                            Copy from /gdata to /gdata
                                            (gluster to gluster)<br>
                                          </b>[root@mseas-data2 gdata]#
                                          dd if=/gdata/zero1
                                          of=/gdata/zero2<br>
                                          2048000+0 records in<br>
                                          2048000+0 records out<br>
                                          1048576000 bytes (1.0 GB)
                                          copied, 101.052 s, <b>10.4
                                            MB/s</b> - realllyyy
                                          slooowww<br>
                                          <br>
                                          <br>
                                          <b>Copy from /gdata to /gdata</b>
                                          <b>2nd time <b>(gluster to
                                              gluster)</b></b><br>
                                          [root@mseas-data2 gdata]# dd
                                          if=/gdata/zero1
                                          of=/gdata/zero2<br>
                                          2048000+0 records in<br>
                                          2048000+0 records out<br>
                                          1048576000 bytes (1.0 GB)
                                          copied, 92.4904 s, <b>11.3
                                            MB/s</b> <span>- realllyyy
                                            slooowww</span> again<br>
                                          <br>
                                          <br>
                                          <br>
                                          <b>Copy from /home to /home
                                            (ext4 to ext4)</b><br>
                                          [root@mseas-data2 gdata]# dd
                                          if=/home/zero1 of=/home/zero2<br>
                                          2048000+0 records in<br>
                                          2048000+0 records out<br>
                                          1048576000 bytes (1.0 GB)
                                          copied, 3.53263 s, <b>297
                                            MB/s </b>30 times as fast<br>
                                          <br>
                                          <br>
                                          <b>Copy from /home to /home
                                            (ext4 to ext4)</b><br>
                                          [root@mseas-data2 gdata]# dd
                                          if=/home/zero1 of=/home/zero3<br>
                                          2048000+0 records in<br>
                                          2048000+0 records out<br>
                                          1048576000 bytes (1.0 GB)
                                          copied, 4.1737 s, <b>251 MB/s</b>
                                          <span>- 30 times as fast<br>
                                            <br>
                                            <br>
                                            As a test, can we copy data
                                            directly to the xfs
                                            mountpoint (/mnt/brick1) and
                                            bypass gluster?<br>
                                            <br>
                                            <br>
                                            Any help you could give us
                                            would be appreciated.<br>
                                            <br>
                                          </span>Thanks<br>
                                        </p>
                                        <pre class="m_-8630651799581052987gmail-m_1278894059907384689moz-signature" cols="72">-- 

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=<wbr>-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=<wbr>-=-=-
Pat Haley                          Email:  <a moz-do-not-send="true" class="m_-8630651799581052987gmail-m_1278894059907384689moz-txt-link-abbreviated" href="mailto:phaley@mit.edu" target="_blank">phaley@mit.edu</a>
Center for Ocean Engineering       Phone:  (617) 253-6824
Dept. of Mechanical Engineering    Fax:    (617) 253-8125
MIT, Room 5-213                    <a moz-do-not-send="true" class="m_-8630651799581052987gmail-m_1278894059907384689moz-txt-link-freetext" href="http://web.mit.edu/phaley/www/" target="_blank">http://web.mit.edu/phaley/www/</a>
77 Massachusetts Avenue
Cambridge, MA  02139-4301
</pre>
      

      <fieldset class="m_-8630651799581052987gmail-m_1278894059907384689mimeAttachmentHeader"></fieldset>
      

      </div></div><pre>______________________________<wbr>_________________
Gluster-users mailing list
<a moz-do-not-send="true" class="m_-8630651799581052987gmail-m_1278894059907384689moz-txt-link-abbreviated" href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a>
<a moz-do-not-send="true" class="m_-8630651799581052987gmail-m_1278894059907384689moz-txt-link-freetext" href="http://lists.gluster.org/mailman/listinfo/gluster-users" target="_blank">http://lists.gluster.org/mailm<wbr>an/listinfo/gluster-users</a></pre>
    </blockquote>
    <p>

    </p>
  </div>


______________________________<wbr>_________________

Gluster-users mailing list

<a moz-do-not-send="true" href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a>

<a moz-do-not-send="true" href="http://lists.gluster.org/mailman/listinfo/gluster-users" rel="noreferrer" target="_blank">http://lists.gluster.org/mailm<wbr>an/listinfo/gluster-users</a>
</blockquote></div>


-- 
<div class="m_-8630651799581052987gmail_signature"><div dir="ltr">Pranith
</div></div>
</div></div>



</blockquote><p>
</p>


</blockquote>
<pre class="m_-8630651799581052987moz-signature" cols="72">-- 

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=<wbr>-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=<wbr>-=-=-
Pat Haley                          Email:  <a moz-do-not-send="true" class="m_-8630651799581052987moz-txt-link-abbreviated" href="mailto:phaley@mit.edu" target="_blank">phaley@mit.edu</a>
Center for Ocean Engineering       Phone:  (617) 253-6824
Dept. of Mechanical Engineering    Fax:    (617) 253-8125
MIT, Room 5-213                    <a moz-do-not-send="true" class="m_-8630651799581052987moz-txt-link-freetext" href="http://web.mit.edu/phaley/www/" target="_blank">http://web.mit.edu/phaley/www/</a>
77 Massachusetts Avenue
Cambridge, MA  02139-4301
</pre></div></div></div></blockquote></div>


-- 
<div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr">Pranith
</div></div>
</div>



</blockquote>
<pre class="moz-signature" cols="72">-- 

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Pat Haley                          Email:  <a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:phaley@mit.edu">phaley@mit.edu</a>
Center for Ocean Engineering       Phone:  (617) 253-6824
Dept. of Mechanical Engineering    Fax:    (617) 253-8125
MIT, Room 5-213                    <a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://web.mit.edu/phaley/www/">http://web.mit.edu/phaley/www/</a>
77 Massachusetts Avenue
Cambridge, MA  02139-4301
</pre>


</blockquote><p>
</p></body></html>