<div dir="ltr">Hi,<br><div><br>Speaking from shard translator's POV, one thing you can do to improve performance is to use preallocated images.<br>This
will at least eliminate the need for shard to perform multiple steps as
part of the writes - such as creating the shard and then writing to it
and then updating the aggregated file size - all of which require one
network call each, which further get blown up once they reach AFR
(replicate) into many more network calls. What this also means is that the performance with and without shard will be the same with this change.<br><br><div>Also, could you also enable client-io-threads and see if that improves performance?<br><br></div>There's a patch that is part of 3.11.1 that has been found to improve performance for vm workloads based on our testing - <a href="https://review.gluster.org/#/c/17391/">https://review.gluster.org/#/c/17391/</a><br>You can give this version a try.<br><br>-Krutika</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Sep 4, 2017 at 7:48 PM, Roei G <span dir="ltr"><<a href="mailto:ganor.roei98@gmail.com" target="_blank">ganor.roei98@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"><div><div class="gmail_quote"><blockquote class="m_2713354091771095627quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto">Hey everyone!<div dir="auto">I have deployed gluster on 3 nodes with 4 SSDs each and 10Gb Ethernet connection.</div><div dir="auto"><br></div><div dir="auto">The storage is configured with 3 gluster volumes, every volume has 12 bricks (4 bricks on every server, 1 per ssd in the server).</div><div dir="auto"><br></div><div dir="auto">With the 'features.shard' off option my writing speed (using the 'dd' command) is approximately 250 Mbs and when the feature is on the writing speed is around 130mbs.</div><div dir="auto"><br></div><div dir="auto">--------- gluster version 3.8.13 --------</div><div dir="auto"><br></div><div dir="auto">Volume name: data</div><div dir="auto">Number of bricks : 4 * 3 = 12</div><div dir="auto">Bricks:</div><div dir="auto">Brick1: server1:/brick/data1</div><div dir="auto">Brick2: server1:/brick/data2</div><div dir="auto">Brick3: server1:/brick/data3</div><div dir="auto">Brick4: server1:/brick/data4</div><div dir="auto">Brick5: server2:/brick/data1</div><div dir="auto">.</div><div dir="auto">.</div><div dir="auto">.</div><div dir="auto">Options reconfigure:</div><div dir="auto">Performance.strict-o-direct: off</div><div dir="auto">Cluster.nufa: off</div><div dir="auto">Features.shard-block-size: 512MB</div><div dir="auto">Features.shard: on</div><div dir="auto">Cluster.server-quorum-type: server</div><div dir="auto">Cluster.quorum-type: auto</div><div dir="auto">Cluster.eager-lock: enable</div><div dir="auto">Network.remote-dio: on</div><div dir="auto">Performance.readdir-ahead: on</div><div dir="auto"><br></div><div dir="auto">Any idea on how to improve my performance?</div><div dir="auto"><br></div></div>
</blockquote></div><br></div></div>
<br>______________________________<wbr>_________________<br>
Gluster-users mailing list<br>
<a href="mailto:Gluster-users@gluster.org">Gluster-users@gluster.org</a><br>
<a href="http://lists.gluster.org/mailman/listinfo/gluster-users" rel="noreferrer" target="_blank">http://lists.gluster.org/<wbr>mailman/listinfo/gluster-users</a><br></blockquote></div><br></div>