<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <p><font face="Helvetica, Arial, sans-serif">Hello,</font></p>
    <p><font face="Helvetica, Arial, sans-serif">Gluster-Performance is
        bad. Thats why I asked for native qemu-libgfapi access for
        Ovirt-VM's to gluster volumes which I thought to be possible
        since 3.6.x. Documentation is misleading and still in 4.1.2
        Ovirt is using fuse to mount gluster-based VM-Disks.</font></p>
    <p><font face="Helvetica, Arial, sans-serif"></font>Bye<br>
    </p>
    <br>
    <div class="moz-cite-prefix">Am 19.06.2017 um 17:23 schrieb Darrell
      Budic:<br>
    </div>
    <blockquote type="cite"
      cite="mid:D8AAF9DB-02FB-4B38-9E33-174134F5377C@onholyground.com">
      <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
      Chris-
      <div class=""><br class="">
      </div>
      <div class="">You probably need to head over to <a
          href="mailto:gluster-users@gluster.org" class=""
          moz-do-not-send="true">gluster-users@gluster.org</a> for help
        with performance issues.</div>
      <div class=""><br class="">
      </div>
      <div class="">That said, what kind of performance are you getting,
        via some form or testing like bonnie++ or even dd runs? Raw
        bricks vs gluster performance is useful to determine what kind
        of performance you’re actually getting.</div>
      <div class=""><br class="">
      </div>
      <div class="">Beyond that, I’d recommend dropping the arbiter
        bricks and re-adding them as full replicas, they can’t serve
        distributed data in this configuration and may be slowing things
        down on you. If you’ve got a storage network setup, make sure
        it’s using the largest MTU it can, and consider adding/testing
        these settings that I use on my main storage volume:</div>
      <div class=""><br class="">
      </div>
      <div class="">
        <div style="margin: 0px; line-height: normal;" class=""><a
            href="http://performance.io" class="" moz-do-not-send="true">performance.io</a>-thread-count:
          32</div>
        <div style="margin: 0px; line-height: normal;" class=""><span
            style="font-variant-ligatures: no-common-ligatures" class="">client.event-threads:
            8</span></div>
        <div style="margin: 0px; line-height: normal;" class=""><span
            style="font-variant-ligatures: no-common-ligatures" class="">server.event-threads:
            3</span></div>
        <div style="margin: 0px; line-height: normal;" class="">performance.stat-prefetch:
          on</div>
      </div>
      <div class=""><span style="font-variant-ligatures:
          no-common-ligatures" class=""><br class="">
        </span></div>
      <div class=""><span style="font-variant-ligatures:
          no-common-ligatures" class="">Good luck,</span></div>
      <div class=""><span style="font-variant-ligatures:
          no-common-ligatures" class=""><br class="">
        </span></div>
      <div class=""><span style="font-variant-ligatures:
          no-common-ligatures" class="">  -Darrell</span></div>
      <div class=""><br class="">
      </div>
      <div class=""><br class="">
        <div>
          <blockquote type="cite" class="">
            <div class="">On Jun 19, 2017, at 9:46 AM, Chris Boot &lt;<a
                href="mailto:bootc@bootc.net" class=""
                moz-do-not-send="true">bootc@bootc.net</a>&gt; wrote:</div>
            <br class="Apple-interchange-newline">
            <div class="">
              <div class="">Hi folks,<br class="">
                <br class="">
                I have 3x servers in a "hyper-converged" oVirt 4.1.2 +
                GlusterFS 3.10<br class="">
                configuration. My VMs run off a replica 3 arbiter 1
                volume comprised of<br class="">
                6 bricks, which themselves live on two SSDs in each of
                the servers (one<br class="">
                brick per SSD). The bricks are XFS on LVM thin volumes
                straight onto the<br class="">
                SSDs. Connectivity is 10G Ethernet.<br class="">
                <br class="">
                Performance within the VMs is pretty terrible. I
                experience very low<br class="">
                throughput and random IO is really bad: it feels like a
                latency issue.<br class="">
                On my oVirt nodes the SSDs are not generally very busy.
                The 10G network<br class="">
                seems to run without errors (iperf3 gives bandwidth
                measurements of &gt;=<br class="">
                9.20 Gbits/sec between the three servers).<br class="">
                <br class="">
                To put this into perspective: I was getting better
                behaviour from NFS4<br class="">
                on a gigabit connection than I am with GlusterFS on 10G:
                that doesn't<br class="">
                feel right at all.<br class="">
                <br class="">
                My volume configuration looks like this:<br class="">
                <br class="">
                Volume Name: vmssd<br class="">
                Type: Distributed-Replicate<br class="">
                Volume ID: d5a5ddd1-a140-4e0d-b514-701cfe464853<br
                  class="">
                Status: Started<br class="">
                Snapshot Count: 0<br class="">
                Number of Bricks: 2 x (2 + 1) = 6<br class="">
                Transport-type: tcp<br class="">
                Bricks:<br class="">
                Brick1: ovirt3:/gluster/ssd0_vmssd/brick<br class="">
                Brick2: ovirt1:/gluster/ssd0_vmssd/brick<br class="">
                Brick3: ovirt2:/gluster/ssd0_vmssd/brick (arbiter)<br
                  class="">
                Brick4: ovirt3:/gluster/ssd1_vmssd/brick<br class="">
                Brick5: ovirt1:/gluster/ssd1_vmssd/brick<br class="">
                Brick6: ovirt2:/gluster/ssd1_vmssd/brick (arbiter)<br
                  class="">
                Options Reconfigured:<br class="">
                nfs.disable: on<br class="">
                transport.address-family: inet6<br class="">
                performance.quick-read: off<br class="">
                performance.read-ahead: off<br class="">
                <a href="http://performance.io" class=""
                  moz-do-not-send="true">performance.io</a>-cache: off<br
                  class="">
                performance.stat-prefetch: off<br class="">
                performance.low-prio-threads: 32<br class="">
                network.remote-dio: off<br class="">
                cluster.eager-lock: enable<br class="">
                cluster.quorum-type: auto<br class="">
                cluster.server-quorum-type: server<br class="">
                cluster.data-self-heal-algorithm: full<br class="">
                cluster.locking-scheme: granular<br class="">
                cluster.shd-max-threads: 8<br class="">
                cluster.shd-wait-qlength: 10000<br class="">
                features.shard: on<br class="">
                user.cifs: off<br class="">
                storage.owner-uid: 36<br class="">
                storage.owner-gid: 36<br class="">
                features.shard-block-size: 128MB<br class="">
                performance.strict-o-direct: on<br class="">
                network.ping-timeout: 30<br class="">
                cluster.granular-entry-heal: enable<br class="">
                <br class="">
                I would really appreciate some guidance on this to try
                to improve things<br class="">
                because at this rate I will need to reconsider using
                GlusterFS altogether.<br class="">
                <br class="">
                Cheers,<br class="">
                Chris<br class="">
                <br class="">
                -- <br class="">
                Chris Boot<br class="">
                <a href="mailto:bootc@bootc.net" class=""
                  moz-do-not-send="true">bootc@bootc.net</a><br class="">
                _______________________________________________<br
                  class="">
                Users mailing list<br class="">
                <a class="moz-txt-link-abbreviated" href="mailto:Users@ovirt.org">Users@ovirt.org</a><br class="">
                <a class="moz-txt-link-freetext" href="http://lists.ovirt.org/mailman/listinfo/users">http://lists.ovirt.org/mailman/listinfo/users</a><br
                  class="">
              </div>
            </div>
          </blockquote>
        </div>
        <br class="">
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
Users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Users@ovirt.org">Users@ovirt.org</a>
<a class="moz-txt-link-freetext" href="http://lists.ovirt.org/mailman/listinfo/users">http://lists.ovirt.org/mailman/listinfo/users</a>
</pre>
    </blockquote>
    <br>
    <div class="moz-signature">-- <br>
      <p>
      </p>
      <table cellspacing="0" cellpadding="0" border="0">
        <tbody>
          <tr>
            <td colspan="3"><img
                src="cid:part6.276D40AB.8385CD25@databay.de" height="30"
                width="151" border="0"></td>
          </tr>
          <tr>
            <td valign="top"> <font size="-1" face="Verdana, Arial,
                sans-serif"><br>
                <b>Ralf Schenk</b><br>
                fon +49 (0) 24 05 / 40 83 70<br>
                fax +49 (0) 24 05 / 40 83 759<br>
                mail <a href="mailto:rs@databay.de"><font
                    color="#FF0000"><b>rs@databay.de</b></font></a><br>
              </font> </td>
            <td width="30"> </td>
            <td valign="top"> <font size="-1" face="Verdana, Arial,
                sans-serif"><br>
                <b>Databay AG</b><br>
                Jens-Otto-Krag-Straße 11<br>
                D-52146 Würselen<br>
                <a href="http://www.databay.de"><font color="#FF0000"><b>www.databay.de</b></font></a>
              </font> </td>
          </tr>
          <tr>
            <td colspan="3" valign="top"> <font size="1" face="Verdana,
                Arial, sans-serif"><br>
                Sitz/Amtsgericht Aachen • HRB:8437 • USt-IdNr.: DE
                210844202<br>
                Vorstand: Ralf Schenk, Dipl.-Ing. Jens Conze, Aresch
                Yavari, Dipl.-Kfm. Philipp Hermanns<br>
                Aufsichtsratsvorsitzender: Wilhelm Dohmen </font> </td>
          </tr>
        </tbody>
      </table>
      <hr noshade="noshade" size="1" color="#000000" width="100%">
    </div>
  </body>
</html>