<div dir="ltr"><div>Dear Gluster Community,</div><div><br></div><div>it is a known fact that Samba+Gluster has a bad smallfile performance. We now have some test measurements created by this setup: 2-Node-Cluster on real hardware with Replica-2 Volume (just one subvolume), Gluster v.4.1.6, Samba v4.7. Samba writes to Gluster via FUSE. Files created by fio. We used a Windows System as Client which is in the same network like the servers.<br></div><div><br></div><div>The measurements are as follows. In each test case 400 files were written:</div><div><br></div><div> 64KiB_x_400 files 1MiB_x_400 files 10MiB_x_400 files<br></div><div>1 Thread 0,77 MiB/s 8,05 MiB/s 72,67 MiB/s<br></div><div>4 Threads 0,86 MiB/s 8,92 MiB/s 90,38 MiB/s <br></div><div>8 Threads 0,87 MiB/s 8,92 MiB/s 94,75 MiB/s</div><div><br></div><div>
<pre class="gmail-tw-data-text gmail-tw-ta gmail-tw-text-small" id="gmail-tw-target-text" style="text-align:left;height:216px" dir="ltr"><span style="font-family:arial,helvetica,sans-serif"><span lang="en">Does anyone have measurements that are in a similar range or are significantly different?<br>We do not know which values can still be considered "normal" and which are not. <br>We also know that there are options to improve performance. But first of all we are interested <br>in whether there are reference values.<br>
<span style="font-family:arial,helvetica,sans-serif"><span lang="en">Regards<br>David Spisla</span></span></span></span></pre></div></div>