<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div></div><div>Can you please post your logfile otherwise it’s hard to tell what’s going wrong.</div><div><br>On 20 Sep 2017, at 12:20, "<a href="mailto:nek5000-users@lists.mcs.anl.gov">nek5000-users@lists.mcs.anl.gov</a>" <<a href="mailto:nek5000-users@lists.mcs.anl.gov">nek5000-users@lists.mcs.anl.gov</a>> wrote:<br><br></div><blockquote type="cite"><div><div dir="ltr"><div>Dear Nek devs<br><br>Usually I start some calculation with 100-150 processors and it runs with acceptable speed. But sometimes the speed of calculation is extremely low, something like 1 step per 10 hours with 50 million mesh points. It seems like all computational nodes fall on one processors, while the rest just stay without any job, despite that all requirements about nodes per processor are satisfied.<br><br></div><div>Have you ever seen something like that ? And how it can be fixed ?<br></div><div><br></div>Regards, Vatslav<br clear="all">
</div>
<pre>_______________________________________________<br>
Nek5000-users mailing list<br>
<a href="mailto:Nek5000-users@lists.mcs.anl.gov">Nek5000-users@lists.mcs.anl.gov</a><br>
<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users">https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users</a><br>
</pre></div></blockquote></body></html>