<div dir="ltr">Leo,<div><br></div><div>NFS can be a hindrance but if tuned and configured properly might not be as terrible. Some thoughts...</div><div><ul><li>What interface are the nodes accessing NFS via? Ethernet or Infiniband?</li><li>Have you tuned the number of NFS server threads above defaults? </li><li>As a test, you could deploy a single Lustre node that would act as MGS/MDS and OSS simultaneously to test for performance gains via Infiniband. </li><li>Your scratch volume must really be scratch because you are running with no parity protection (two disk os SSD stripe)</li><li>You're probably better off with tuned NFS as opposed to GlusterFS</li></ul><div>--Jeff</div></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Aug 10, 2023 at 12:19 PM leo camilo <<a href="mailto:lhcamilo@gmail.com">lhcamilo@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><div><div><div><div><div><div>Hi everyone, <br><br></div>I was hoping I would seek some sage advice from you guys. <br><br></div>At my department we have build this small prototyping cluster with 5 compute nodes,1 name node and 1 file server. <br><br></div>Up until now, the name node contained the scratch partition, which consisted of 2x4TB HDD, which form an 8 TB striped zfs pool. The pool is shared to all the nodes using nfs. The compute nodes and the name node and compute nodes are connected with both cat6 ethernet net cable and infiniband. Each compute node has 40 cores.<br><br></div>Recently I have attempted to launch computation from each node (40 tasks per node), so 1 computation per node. And the performance was abysmal. I reckon I might have reached the limits of NFS.<br><br></div>I then realised that this was due to very poor performance from NFS. I am not using stateless nodes, so each node has about 200 GB of SSD storage and running directly from there was a lot faster. <br><br></div>So, to solve the issue, I reckon I should replace NFS with something better. I have ordered 2x4TB NVMEs for the new scratch and I was thinking of :<br><br></div><ul><li>using the 2x4TB NVME in a striped ZFS pool and use a single node GlusterFS to replace NFS</li><li>using the 2x4TB NVME with GlusterFS in a distributed arrangement (still single node)</li></ul><div>Some people told me to use lustre,but I reckon that might be overkill. And I would only use a single fileserver machine(1 node).<br><br></div><div>Could you guys give me some sage advice here?<br><br></div><div>Thanks in advance<br></div><div><div><div><div><div><br><br></div></div></div></div></div></div>
_______________________________________________<br>
Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org" target="_blank">Beowulf@beowulf.org</a> sponsored by Penguin Computing<br>
To change your subscription (digest mode or unsubscribe) visit <a href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf" rel="noreferrer" target="_blank">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a><br>
</blockquote></div><br clear="all"><div><br></div><span class="gmail_signature_prefix">-- </span><br><div dir="ltr" class="gmail_signature"><div dir="ltr"><div><div dir="ltr">------------------------------<br>Jeff Johnson<br>Co-Founder<br>Aeon Computing<br><br><a href="mailto:jeff.johnson@aeoncomputing.com" target="_blank">jeff.johnson@aeoncomputing.com</a><br><a href="http://www.aeoncomputing.com" target="_blank">www.aeoncomputing.com</a><br>t: 858-412-3810 x1001 f: 858-412-3845<br>m: 619-204-9061<br><br>4170 Morena Boulevard, Suite C - San Diego, CA 92117<div><br></div><div>High-Performance Computing / Lustre Filesystems / Scale-out Storage</div></div></div></div></div>