Hi Matt,<br>
As you config file, client and server are located at the same
machine , and they use tcp socket for communication, unix socket may be some improvement in performance .<br>
<br>
:D<br>
<br>
Thanks<br>
Baggio<br>
<br><br><div class="gmail_quote">2008/6/13 Daniel Jordan Bambach <<a href="mailto:dan@lateral.net" target="_blank">dan@lateral.net</a>>:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Heya Matt,<br>
<br>
as per Anands mail to me on the 5th June, much performance can be gained by combining the server definitions into the client config and not running glusterfsd at all.<br>
<br>
I set this up and saw some improvements, but unfortunately there is still an outstanding issue about getting the two machines to connect to each other properly - I was hoping that Anand might be able to help out on that, but he doesn't seem to have been around recently.<br>
<font color="#888888">
<br>
D.</font><div><div></div><div><br>
<br>
<br>
On 12 Jun 2008, at 17:33, Matt Brown wrote:<br>
<br>
</div></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div><div></div><div>
On Wednesday 11 June 2008 06:04:16 pm baggio liu wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi Matt:<br>
Can I have the detail comfig of your cluster?? like node number,<br>
config of client and server ?? and so on.<br>
</blockquote>
<br>
I have two machines, both acting as clients & servers. Below are the fully<br>
loaded configs, using server side unify over afr with various performance<br>
translators. As you'll see I'm using glusterfs for the /home dir. In users<br>
home directories I have php based websites that are being served by apache .<br>
The loading of the web pages is very slow compared to using non-gluster file<br>
systems on the same machines. I was concerned that perhaps using glusterfs on<br>
top of software raid might be the issue. But if that's known to *not*<br>
significantly impact performance, then perhaps I should be looking at another<br>
way of building this config? Thanks for any insight.<br>
Matt<br>
<br>
# machine 1 server config<br>
volume home-web1-ds<br>
type storage/posix<br>
option directory /opt/gluster/home<br>
end-volume<br>
<br>
volume home-web1-ns<br>
type storage/posix<br>
option directory /opt/gluster/home-ns<br>
end-volume<br>
<br>
volume home-web2-ds<br>
type protocol/client<br>
option transport-type tcp/client<br>
option remote-host <a href="http://10.0.3.22" target="_blank">10.0.3.22</a><br>
option remote-subvolume home-web2-ds<br>
end-volume<br>
<br>
volume home-web2-ns<br>
type protocol/client<br>
option transport-type tcp/client<br>
option remote-host <a href="http://10.0.3.22" target="_blank">10.0.3.22</a><br>
option remote-subvolume home-web2-ns<br>
end-volume<br>
<br>
volume home-ns-afr<br>
type cluster/afr<br>
option read-subvolume home-web1-ns<br>
subvolumes home-web1-ns home-web2-ns<br>
end-volume<br>
<br>
volume home-ds-afr<br>
type cluster/afr<br>
option read-subvolume home-web1-ds<br>
subvolumes home-web1-ds home-web2-ds<br>
end-volume<br>
<br>
volume home-unify<br>
type cluster/unify<br>
subvolumes home-ds-afr<br>
option namespace home-ns-afr<br>
option scheduler rr<br>
end-volume<br>
<br>
volume home<br>
type performance/io-threads<br>
option thread-count 2<br>
option cache-size 64MB<br>
subvolumes home-unify<br>
end-volume<br>
<br>
volume server<br>
type protocol/server<br>
option transport-type tcp/server<br>
subvolumes home<br>
option auth.ip.home-web1-ds.allow <a href="http://10.0.3." target="_blank">10.0.3.</a>*,<a href="http://127.0.0.1" target="_blank">127.0.0.1</a><br>
option auth.ip.home-web1-ns.allow <a href="http://10.0.3." target="_blank">10.0.3.</a>*,<a href="http://127.0.0.1" target="_blank">127.0.0.1</a><br>
option auth.ip.home.allow <a href="http://10.0.3." target="_blank">10.0.3.</a>*,<a href="http://127.0.0.1" target="_blank">127.0.0.1</a><br>
end-volume<br>
<br>
# machine 1 client config<br>
volume home-mount<br>
type protocol/client<br>
option transport-type tcp/client<br>
option remote-host <a href="http://127.0.0.1" target="_blank">127.0.0.1</a><br>
option remote-subvolume home<br>
end-volume<br>
<br>
volume ioc<br>
type performance/io-cache<br>
subvolumes home-mount<br>
option page-size 1MB<br>
option cache-size 128MB<br>
option force-revalidate-timeout 10<br>
end-volume<br>
<br>
volume writeback<br>
type performance/write-behind<br>
option aggregate-size 1MB<br>
subvolumes ioc<br>
end-volume<br>
<br>
volume readahead<br>
type performance/read-ahead<br>
option page-size 1MB<br>
option page-count 4<br>
option force-atime-update no<br>
subvolumes writeback<br>
end-volume<br>
<br>
<br>
# server 2 server config<br>
volume home-web2-ds<br>
type storage/posix<br>
option directory /opt/gluster/home<br>
end-volume<br>
<br>
volume home-web2-ns<br>
type storage/posix<br>
option directory /opt/gluster/home-ns<br>
end-volume<br>
<br>
volume home-web1-ds<br>
type protocol/client<br>
option transport-type tcp/client<br>
option remote-host <a href="http://10.0.3.21" target="_blank">10.0.3.21</a><br>
option remote-subvolume home-web1-ds<br>
end-volume<br>
<br>
volume home-web1-ns<br>
type protocol/client<br>
option transport-type tcp/client<br>
option remote-host <a href="http://10.0.3.21" target="_blank">10.0.3.21</a><br>
option remote-subvolume home-web1-ns<br>
end-volume<br>
<br>
volume home-ns-afr<br>
type cluster/afr<br>
option read-subvolume home-web2-ns<br>
subvolumes home-web1-ns home-web2-ns<br>
end-volume<br>
<br>
volume home-ds-afr<br>
type cluster/afr<br>
option read-subvolume home-web2-ds<br>
subvolumes home-web1-ds home-web2-ds<br>
end-volume<br>
<br>
volume home-unify<br>
type cluster/unify<br>
subvolumes home-ds-afr<br>
option namespace home-ns-afr<br>
option scheduler rr<br>
end-volume<br>
<br>
volume home<br>
type performance/io-threads<br>
option thread-count 2<br>
option cache-size 64MB<br>
subvolumes home-unify<br>
end-volume<br>
<br>
volume server<br>
type protocol/server<br>
option transport-type tcp/server<br>
subvolumes home<br>
option auth.ip.home-web2-ds.allow <a href="http://10.0.3." target="_blank">10.0.3.</a>*,<a href="http://127.0.0.1" target="_blank">127.0.0.1</a><br>
option auth.ip.home-web2-ns.allow <a href="http://10.0.3." target="_blank">10.0.3.</a>*,<a href="http://127.0.0.1" target="_blank">127.0.0.1</a><br>
option auth.ip.home.allow <a href="http://10.0.3." target="_blank">10.0.3.</a>*,<a href="http://127.0.0.1" target="_blank">127.0.0.1</a><br>
end-volume<br>
<br>
# sever 2 client config<br>
volume home-mount<br>
type protocol/client<br>
option transport-type tcp/client<br>
option remote-host <a href="http://127.0.0.1" target="_blank">127.0.0.1</a><br>
option remote-subvolume home<br>
end-volume<br>
<br>
volume ioc<br>
type performance/io-cache<br>
subvolumes home-mount<br>
option page-size 1MB<br>
option cache-size 128MB<br>
option force-revalidate-timeout 10<br>
end-volume<br>
<br>
volume writeback<br>
type performance/write-behind<br>
option aggregate-size 1MB<br>
subvolumes ioc<br>
end-volume<br>
<br>
volume readahead<br>
type performance/read-ahead<br>
option page-size 1MB<br>
option page-count 4<br>
option force-atime-update no<br>
subvolumes writeback<br>
end-volume<br>
<br></div></div><div>
_______________________________________________<br>
Gluster-users mailing list<br>
<a href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a><br>
<a href="http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users" target="_blank">http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users</a><br>
</div></blockquote>
<br>
</blockquote></div><br>