<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Jan 11, 2018 at 10:44 PM, Darrell Budic <span dir="ltr"><<a href="mailto:budic@onholyground.com" target="_blank">budic@onholyground.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div style="overflow-wrap: break-word;">Sounds like a good option to look into, but I wouldn’t want it to take time & resources away from other, non-GPU based, methods of improving this. Mainly because I don’t have discrete GPUs in most of my systems. While I could add them to my main server cluster pretty easily, many of my clients are 1U or blade systems and have no real possibility of having a GPU added.</div></blockquote><div><br></div><div>In 3.9.0 Xavi added support to use hardware extensions on CPU if available. It uses special instructions if any of "x64", "sse", "avx" etc is available. There is also a plan to do systematic encoding in future which will be different from the current way of encoding. We need to check how performance will change with that as well.<br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div style="overflow-wrap: break-word;"><div><br></div><div>It would also add physical resource requirements to future client deploys, requiring more than 1U for the server (most likely), and I’m not likely to want to do this if I’m trying to optimize for client density, especially with the cost of GPUs today. <br></div></div></blockquote><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div style="overflow-wrap: break-word;"><div><div><br><div><blockquote type="cite"><hr style="border-color:rgb(181,196,223) currentcolor currentcolor;border-style:solid none none;border-width:1pt medium medium;padding:0px;margin:10px 0px 5px"><span id="gmail-m_-1082143859864860373RwhHeaderAttributes"><font style="font:normal normal normal normal 13px Helvetica;color:rgb(0,0,0)" size="4" color="#000000" face="Helvetica"><b>From:</b> Ashish Pandey <<a href="mailto:aspandey@redhat.com" target="_blank">aspandey@redhat.com</a>></font></span><br>
<span><font style="font:normal normal normal normal 13px Helvetica;color:rgb(0,0,0)" size="4" color="#000000" face="Helvetica"><b>Subject:</b> [Gluster-users] Integration of GPU with glusterfs</font></span><br>
<span><font style="font:normal normal normal normal 13px Helvetica;color:rgb(0,0,0)" size="4" color="#000000" face="Helvetica"><b>Date:</b> January 11, 2018 at 12:12:43 AM CST</font></span><br>
<span><font style="font:normal normal normal normal 13px Helvetica;color:rgb(0,0,0)" size="4" color="#000000" face="Helvetica"><b>To:</b> Gluster Users</font></span><br>
<span><font style="font:normal normal normal normal 13px Helvetica;color:rgb(0,0,0)" size="4" color="#000000" face="Helvetica"><b>Cc:</b> Gluster Devel</font></span><br>
<br class="gmail-m_-1082143859864860373Apple-interchange-newline"><div><div><div class="gmail-h5"><div><div style="font-family:"times new roman","new york",times,serif;font-size:12pt"><div>Hi,<br></div><div><br></div><div>We have been thinking of exploiting GPU capabilities to enhance performance of glusterfs. We would like to know others thoughts on this.<br></div><div>In EC, we have been doing CPU intensive computations to encode and decode data before writing and reading. This requires a lot of CPU cycles and we have<br></div><div>been observing 100% CPU usage on client side. Data healing will also have the same impact as it also needs to do read-decode-encode-write cycle. </div><div>As most of the modern servers comes with GPU feature, having glusterfs GPU ready might give us performance improvements.<br></div><div>This is not only specific to EC volume, there are other features which will require a lot of computations and could use this capability; For Example:<br></div><div>1 - Encryption/Decryption<br></div><div>2 - Compression and de-duplication<br></div><div>3 - Hashing<br></div><div>4 - Any other? [Please add if you have something in mind]<br></div><div><br></div><div>Before proceeding further we would like to have your inputs on this.</div><div>Do you have any other use case (existing or future) which could perform better on GPU?<br></div><div>Do you think that it is worth to integrate GPU with glusterfs? The effort to have this performance gain could be achieved by some other better ways.<br></div><div>Any input on the way we should implement it.<br></div><div><br></div><div>There is a gihub issue opened for this. Please provide your comment or reply to this mail.<br></div><div><br></div><div>A - <a href="https://github.com/gluster/glusterfs/issues/388" target="_blank">https://github.com/gluster/<wbr>glusterfs/issues/388</a> <br></div><div><br></div><div>---<br></div><div>Ashish<br></div></div></div></div></div>______________________________<wbr>_________________<br>Gluster-users mailing list<br><a href="mailto:Gluster-users@gluster.org" target="_blank">Gluster-users@gluster.org</a><br><a href="http://lists.gluster.org/mailman/listinfo/gluster-users" target="_blank">http://lists.gluster.org/<wbr>mailman/listinfo/gluster-users</a></div></blockquote></div><br></div></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><br clear="all"><br>-- <br><div class="gmail_signature"><div dir="ltr">Pranith<br></div></div>
</div></div>