Thank you Chris for your help. <br><br>It does exactly what I need!<br><br>Regards,<br><br clear="all"><br>-- <br>Leandro Tavares Carneiro<br>Analista de Suporte Linux/Unix
<br><br><div><span class="gmail_quote">On 2/28/07, <b class="gmail_sendername">Chris Samuel</b> <<a href="mailto:csamuel@vpac.org">csamuel@vpac.org</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
On Wed, 14 Feb 2007, Leandro wrote:<br><br>> What I need is choose nodes by CPUs in a simpler way than "-l<br>> nodes=X:ppn=Y" way. What I'm looking for is a "-l cpus=X" and let the<br>> resource manager/scheduler choose the nodes, packing the tasks in the nodes
<br>> with more CPUs.<br><br>-l nodes=X<br><br>That gets treated as a request for X CPUs anywhere, as long as X < number of<br>nodes. If you want to do X > number of nodes then you need to set the number<br>of nodes in the PBS server by doing:
<br><br>set server resources_available.nodect = X<br><br>How's that ?<br><br>cheers,<br>Chris (way behind with email)<br>--<br> Christopher Samuel - (03)9925 4751 - VPAC Deputy Systems Manager<br> Victorian Partnership for Advanced Computing
<a href="http://www.vpac.org/">http://www.vpac.org/</a><br> Bldg 91, 110 Victoria Street, Carlton South, VIC 3053, Australia<br><br><br>_______________________________________________<br>torqueusers mailing list<br><a href="mailto:torqueusers@supercluster.org">
torqueusers@supercluster.org</a><br><a href="http://www.supercluster.org/mailman/listinfo/torqueusers">http://www.supercluster.org/mailman/listinfo/torqueusers</a><br><br><br></blockquote></div>