<div dir="ltr"><div><div>Hi Vijay,<br><br></div>i deleted the Cluster/Datacenter and set it up with two new (physical) Hosts and now the performance looks great.<br><br></div>I dunno what i did wrong. Thanks a lot....<br></div>
<div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Feb 10, 2014 at 6:10 PM, Vijay Bellur <span dir="ltr"><<a href="mailto:vbellur@redhat.com" target="_blank">vbellur@redhat.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="">On 02/09/2014 11:08 PM, ml ml wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Yes, the only thing which brings the wirte I/O almost on my Host Level<br>
is by enabling viodiskcache = writeback.<br>
As far as i can tell this is caching enabled for the guest and the host<br>
which is critical if sudden power loss happens.<br>
Can i turn this is on if i have a BBU in my Host System?<br>
<br>
</blockquote>
<br></div>
I was referring to the set of gluster volume tunables in [1]. These options can be enabled through "volume set" interface in gluster CLI.<br>
<br>
The quorum options are used for providing tolerance against split-brains and the remaining ones are recommended normally for performance.<br>
<br>
-Vijay<br>
<br>
[1] <a href="https://github.com/gluster/glusterfs/blob/master/extras/group-virt.example" target="_blank">https://github.com/gluster/<u></u>glusterfs/blob/master/extras/<u></u>group-virt.example</a><br>
<br>
</blockquote></div><br></div>