<html><body><div style="font-family: arial, helvetica, sans-serif; font-size: 12pt; color: #000000"><div>Hi everyone,<br></div><div>I've been running some FLASH4.6.2 simulations on Clusters.<br data-mce-bogus="1"></div><div>I noticed that scaling runs with a lot of cores isn't necessarely followed by a speed up of the computation time.</div><div>I was wondering if you have a good way to set up submission file and optimize mpi for FLASH runs on big clusters (nb of threads, cores, processors)?<br data-mce-bogus="1"></div><div><br data-mce-bogus="1"></div><div>Thanks all!<br data-mce-bogus="1"></div><div><br></div><div data-marker="__SIG_POST__">-- <br></div><div><div style="text-align:left"><span style="color:rgb( 0 , 0 , 0 )">Thibault Goudal</span></div><div style="text-align:left"><span style="color:rgb( 0 , 0 , 0 )">PhD Student CEA</span></div><div style="text-align:left"><span style="color:rgb( 0 , 0 , 0 )">CELIA CNRS UMR5107 laboratory</span></div><div style="text-align:left"><span style="color:rgb( 0 , 0 , 0 )">+33664160842</span></div><div style="text-align:left"><span style="color:rgb( 0 , 0 , 0 )">thibault.goudal@u-bordeaux.fr<br></span></div></div></div></body></html>