<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "> I know that I have an almost perfect scaling with 32 H2O going from 1 to 2 cpus.<div><br><div>You have to check what you are measuring (be sure to measure real time).</div><div><br class="webkit-block-placeholder"></div><div>You have to be sure that the mpi correctly sets cpu affinity (so that the process do not swap processors and get slow).</div><div>I know lammpi and openmpi, both do it, I assume mpich also does it.</div><div><br class="webkit-block-placeholder"></div><div>With lammpi you have to write the host filelike this</div><div>myhost cpu=2</div><div><br></div><div>and not</div><div>myhost</div><div>myhost</div><div><br class="webkit-block-placeholder"></div><div>ciao</div><div>Fawzi</div><div>On Jan 25, 2008, at 3:28 PM, Teodoro Laino wrote:</div><div><div><br class="Apple-interchange-newline"><blockquote type="cite"> The problem, unfortunately, is far being so obvious..<div><br class="webkit-block-placeholder"></div><div>I think the MB is just mistyping.. (GB instead..)</div><div>And the problem is not only related to the benchmark input but also to small systems.</div><div>I suggested Luigi to move to the benchmark test since in that case (a larger one) you should definitely see scaling (more difficult</div><div>to observe with smaller systems). IO is reduced to the minimum in testing.. so don't consider the access to the disk.</div><div><br class="webkit-block-placeholder"></div><div>Luigi, I just post the info you gave me about compilers/hardware.. I don't have the compilation flags.. Can you post them here in the group?</div><div><br class="webkit-block-placeholder"></div><div>-------------------------------------------------------------------------------------------------------------------------------------------------</div><div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">Linux k119 2.6.21-1.3194.fc7 #1 SMP Wed May 23 22:47:07 EDT 2007 x86_64 x86_64 x86_64 GNU/Linux</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">gcc version 4.1.2 20070925 (Red Hat 4.1.2-27)</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">ifort Build 20070613 Package ID: l_fc_c_10.0.025</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">intel mkl 10.0.1.014</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">fftw-3.1.2</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">mpich2-1.0.6p1</div><div>-------------------------------------------------------------------------------------------------------------------------------------------------</div></div><div><br></div><div>We use, on our local machines mostly the same setup.. The only difference is the kernel (ours is a 2.6.16.21-0.8-smp, could this be an issue Axel?) and the mpich2 (we use 1.0.5p4)...</div><div><br></div><div>Definitely it's not an OMP issue.. following Manuel suggestion I asked Luigi to set OMP_NUM_THREADS=1 and nothing changes in the timings..</div><div><br class="webkit-block-placeholder"></div><div>Let's see if we can figure out what's going on..</div><div>Sure.. it's a little bit weird...</div><div>Teo</div><div><br class="webkit-block-placeholder"></div><div><div><div>On 25 Jan 2008, at 15:06, Fawzi Mohamed wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0; "><div>well 8MB RAM is for sure a bottleneck ;)</div><div>8GB should not, but, well 6 Hours for 10 steps... maybe it did swap, and if you swap then everything is over, do check, if memory consumption is more than your physical memory the cpu usage drops drastically, again in top you should see this.</div><div><br class="webkit-block-placeholder"></div><div>Use a smaller system, if this is the problem, the 32 water should already scale well...</div></span><br class="Apple-interchange-newline"></blockquote></div><br></div><br> <br> <br></blockquote></div><br></div></div></body></html>