<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">A: One CCL image generating multiple threads via process-run-function.<div class=""><br class=""></div><div class="">This is application code, not an artificial test.</div><div class="">I use a simple scheme where a vector of computations-to-be-done is divided by the number of cores.</div><div class="">The individual computations do not typically run flat out because they share resources, although I have difficulty identifying what resources they might be; improving this is part of my development challenge.</div><div class=""><br class=""></div><div class="">Performance varies up and down, and GC may be one of the causes for reduced parallel performance.</div><div class="">I try to keep GC out of the way: (set-lisp-heap-gc-threshold (* 100 (lisp-heap-gc-threshold)))</div><div class="">I call GC manually between processing stages.</div><div class="">Note that I am running in 110GB of memory and have used only 27 GB.</div><div class=""><br class=""></div><div class="">I’m in uncharted territory, so there is certainly room for improvement.</div><div class="">I am totally impressed that CCL runs out of the box on Knights Landing — congratulations to Clozure Associates !</div><div class=""><br class=""></div><div class="">Martin</div><div class=""><br class=""></div><div class=""><br class=""></div><div class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Jan 30, 2017, at 3:13 PM, Ron Garret <<a href="mailto:ron@flownet.com" class="">ron@flownet.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><meta http-equiv="Content-Type" content="text/html charset=us-ascii" class=""><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">Wow! Cool!<div class=""><br class=""></div><div class="">Is this a single CCL process loading up 100+ cores using native thread, or is it 100+ separate CCL processes? If the former, are you seeing any GC-related performance problems or synchronization issues?</div><div class=""><br class=""></div><div class="">rg</div><div class=""><br class=""><div class=""><div class="">On Jan 30, 2017, at 11:38 AM, martin brooks <<a href="mailto:brooks.martin@sympatico.ca" class="">brooks.martin@sympatico.ca</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite" class=""><meta http-equiv="Content-Type" content="text/html charset=us-ascii" class=""><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">This machine: <a href="http://dap.xeonphi.com/ninja-dev-platform-pedestal-ac.aspx" class="">http://dap.xeonphi.com/ninja-dev-platform-pedestal-ac.aspx</a><div class=""><br class=""></div><div class="">Here is my computational photography code (<a href="http://varilets.org/" class="">varilets.org</a>) generating > 100 core load on 256 cores.</div><div class=""><br class=""></div><div class="">Martin Brooks</div><div class=""><br class=""></div><div class=""><br class=""></div><div class=""><span class=""><100 cores.png></span></div></div>_______________________________________________<br class="">Openmcl-devel mailing list<br class=""><a href="mailto:Openmcl-devel@clozure.com" class="">Openmcl-devel@clozure.com</a><br class=""><a href="https://lists.clozure.com/mailman/listinfo/openmcl-devel" class="">https://lists.clozure.com/mailman/listinfo/openmcl-devel</a><br class=""></blockquote></div><br class=""></div></div></div></blockquote></div><br class=""></div></body></html>