<br><div><span class="gmail_quote">On 1/30/07, <b class="gmail_sendername">Wim Fournier</b> <<a href="mailto:w.fournier@nedstat.nl">w.fournier@nedstat.nl</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;">
Vadivelan Ranjith wrote:<br>><br>><br>> On 1/30/07, *Wim Fournier* <<a href="mailto:w.fournier@nedstat.nl">w.fournier@nedstat.nl</a><br>> <mailto:<a href="mailto:w.fournier@nedstat.nl">w.fournier@nedstat.nl
</a>>> wrote:<br>><br>> Hi all,<br>><br>> I'm still having problems to get the right number of jobs that run<br>> concurrently on each machine.<br>> I just thought, could it be that the default scheduler for torque
<br>> (pbs_sched) doesn't implement this?<br>><br>> Just to make things clear, I want torque to not start more jobs than I<br>> have cores.<br>><br>><br>> Grtz,<br>><br>> Wim Fournier
<br>><br>><br>> Try by Including JO//BNODEMATCHPOLICY EXACTNODE in maui.cfg . //<br>> Send me the output of<br>> qmgr -c 'p s'<br>> and tell me how you are submitting job<br>><br>><br>> Velan
<br>I'm currently using torque without maui.<br>I tried installing maui like half an hour ago, but I'm missing the msub<br>binary and submitting with qsub just leaves the jobs queued.<br><br>Wim<br><br></blockquote>
</div><br>Send me the output of qmgr -c 'p s'. <br>Velan<br>