<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">you can do the following:<div class=""><br class=""></div><div class=""><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class=""># gluster volume set $vol performance.o-thread-count 64</div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class="">Today’s CPU are powerful enough to handle 64 threads per volume.</div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class=""><br class=""></div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class=""># gluster volume set $vol client.event-threads XX</div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class="">XX depend on the number of connections from the FUSE client to the server, you can get this number by running netstat and grep on the server IP and count the number of connections.</div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class=""><br class=""></div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class=""># gluster volume set $vol server.event-threads XX</div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class=""><br class=""></div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class="">XX depend on the number of connections from the server to the client(s), you can get this number by running netstat and grep on “gluster" and count the number of connections.</div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class=""><br class=""></div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class="">also, you can follow the instructions in the following page:</div><div style="margin: 0px; line-height: normal;" class=""><font color="#454545" class=""><a href="http://gluster.readthedocs.org/en/release-3.7.0/Administrator%20Guide/Linux%20Kernel%20Tuning/" class="">http://gluster.readthedocs.org/en/release-3.7.0/Administrator%20Guide/Linux%20Kernel%20Tuning/</a></font></div><div style="margin: 0px; line-height: normal;" class=""><font color="#454545" class=""><br class=""></font></div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class=""><br class=""></div><div style="margin: 0px; line-height: normal; color: rgb(69, 69, 69);" class="">-Bishoy</div><div><blockquote type="cite" class=""><div class="">On Nov 24, 2015, at 1:31 PM, Pierre MGP Pro <<a href="mailto:pierre-mgp-jouy.inra@agsic.net" class="">pierre-mgp-jouy.inra@agsic.net</a>> wrote:</div><br class="Apple-interchange-newline"><div class="">
<meta content="text/html; charset=utf-8" http-equiv="Content-Type" class="">
<div bgcolor="#FFFFFF" text="#000000" class="">
Hi Lindsay Mathieson and all,<br class="">
<br class="">
<br class="">
<div class="moz-cite-prefix">Le 24/11/2015 21:09, Lindsay Mathieson
a écrit :<br class="">
</div>
<blockquote cite="mid:CAEMkAmFHzKebsD=CdcYZ=qEVy1-9pdMbL4PomEBLD8XtPYhCnQ@mail.gmail.com" type="cite" class="">
<div dir="ltr" class="">
<div class="">
<div class="">
<div class="">
<div class="">
<div class="">
<div class="">
<div class="">More details on your setup[ would be useful:<br class="">
</div>
- Number of Nodes<br class="">
</div>
- CPU & RAM on Each Node<br class="">
</div>
- What else is running on the nodes<br class="">
</div>
- Number of bricks<br class="">
</div>
- output of "gluster volume info" & "gluster volume
status"<br class="">
</div>
<br class="">
- ZFS config for each Node<br class="">
</div>
<div class=""> * number of disks and rai arrangement<br class="">
</div>
<div class=""> * log and cache SSD?<br class="">
</div>
* zpool status<br class="">
<br class="">
</div>
</blockquote>
OK I have tested that kind of configuration, and the result depend
of what you are waiting for :<br class="">
<ul class="">
<li class="">zfsonlinux is now efficient, but you will not have access to
the ACL ;</li>
<li class="">on a volume with seven disk we get the maximum of the PCI
Express bandwidth ;</li>
<li class="">so you can mount a distributed gluster volume with your
zfsonlinux nodes. The bandwidth will depend of the kind of
glusterfs volume you want to build : distributed, stripped,
replicated ;</li>
<ul class="">
<li class="">replicated : bad, because of the synchronism write for the
files replication ;</li>
<li class="">striped is the best because it allow you to get an average
bandwidth on a file whatever the node you R/W the file ;</li>
</ul>
<li class="">then the last, for me, is the Ethernet access between each
nodes. If you have 1Gb, get back to your sand box, At the year
you need 10Gbs and as the minimal Ethernet access is two port
you need to bound them ;</li>
<li class="">have 10 Gbs Ethernet switch ;</li>
</ul><p class="">That the expressions of the needs for the now and future
necessities.<br class="">
</p><p class="">sincerely<br class="">
</p><p class="">Pierre Léonard<br class="">
<br class="">
</p>
<br class="">
</div>
_______________________________________________<br class="">Gluster-users mailing list<br class=""><a href="mailto:Gluster-users@gluster.org" class="">Gluster-users@gluster.org</a><br class="">http://www.gluster.org/mailman/listinfo/gluster-users</div></blockquote></div><br class=""></div></body></html>