<div dir="ltr">And I still couldn't find where this large array comes from. </div><div class="gmail_extra"><br><br><div class="gmail_quote">On Sat, Mar 1, 2014 at 3:49 PM, Paolo Giannozzi <span dir="ltr"><<a href="mailto:paolo.giannozzi@uniud.it" target="_blank">paolo.giannozzi@uniud.it</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">On Sat, 2014-03-01 at 14:27 -0600, Peng Chen wrote:<br>
><br>
> And the system is not that large(32 atoms, 400 nband, 8*8*8 kpoints)<br>
> which is run in 128 cores. I think you are probably right that QE is<br>
> trying to allocate a large array somehow.<br>
<br>
</div>... and ?<br>
<div class="HOEnZb"><div class="h5"><br>
> On Fri, Feb 28, 2014 at 10:35 AM, Paolo Giannozzi<br>
> <<a href="mailto:paolo.giannozzi@uniud.it">paolo.giannozzi@uniud.it</a>> wrote:<br>
> On Fri, 2014-02-28 at 09:12 -0600, Peng Chen wrote:<br>
><br>
> > I think it is memory, because the error message is like:<br>
> > : 02/27/2014 14:06:20| main|zeta27|W|job 221982 exceeds job<br>
> hard<br>
> > limit "h_vmem" of queue (2871259136.00000 ><br>
> limit:2147483648.00000) -<br>
> > sending SIGKILL<br>
><br>
><br>
> there are a few hints on how to reduce memory usage to the<br>
> strict<br>
> minimum here:<br>
> <a href="http://www.quantum-espresso.org/wp-content/uploads/Doc/pw_user_guide/node19.html#SECTION000600100000000000000" target="_blank">http://www.quantum-espresso.org/wp-content/uploads/Doc/pw_user_guide/node19.html#SECTION000600100000000000000</a><br>
> If the FFT grid is large, reduce mixing_ndim from its default<br>
> value (8)<br>
> to 4 or so. If the number of bands is large, distribute<br>
> nbnd*nbnd<br>
> matrices using "-ndiag". If you have many k-points, save to<br>
> disk with<br>
> disk_io='medium'. The message you get: "2871259136 ><br>
> limit:<a href="tel:2147483648" value="+12147483648">2147483648</a>"<br>
> makes me think that you crash when trying to allocate an array<br>
> whose<br>
> size is at least 2871259136-2147483648=a lot. It shouldn' be<br>
> difficult<br>
> to figure out where such a large array comes from<br>
><br>
> Paolo<br>
><br>
><br>
> ><br>
> > I normally used h_stak=128M, it is working fine.<br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> > On Fri, Feb 28, 2014 at 7:30 AM, Paolo Giannozzi<br>
> > <<a href="mailto:paolo.giannozzi@uniud.it">paolo.giannozzi@uniud.it</a>> wrote:<br>
> > On Thu, 2014-02-27 at 17:30 -0600, Peng Chen wrote:<br>
> > > P.S. Most of the jobs failed at the beginning of<br>
> scf<br>
> > calculation, and<br>
> > > the length of output scf file is zero.<br>
> ><br>
> ><br>
> > are you sure the problem is the size of the RAM and<br>
> not the<br>
> > size of<br>
> > the stack?<br>
> ><br>
> > P.<br>
> ><br>
> ><br>
> > ><br>
> > ><br>
> > > On Thu, Feb 27, 2014 at 5:09 PM, Peng Chen<br>
> > <<a href="mailto:pchen229@illinois.edu">pchen229@illinois.edu</a>><br>
> > > wrote:<br>
> > > Dear QE users,<br>
> > ><br>
> > ><br>
> > > Recently, our workstation is updated and<br>
> there is a<br>
> > hard limit<br>
> > > on memory (2G per core). Some of QE jobs<br>
> are<br>
> > constantly failed<br>
> > > (not always) because one of the MPI<br>
> processes<br>
> > exceeded the RAM<br>
> > > limit and was killed. I am wondering if<br>
> there is a<br>
> > way to<br>
> > > distribute using memory more evenly in<br>
> every core.<br>
> > ><br>
> > ><br>
> ><br>
> > > _______________________________________________<br>
> > > Pw_forum mailing list<br>
> > > <a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br>
> > > <a href="http://pwscf.org/mailman/listinfo/pw_forum" target="_blank">http://pwscf.org/mailman/listinfo/pw_forum</a><br>
> ><br>
> ><br>
> > --<br>
> > Paolo Giannozzi, Dept.<br>
> Chemistry&Physics&Environment,<br>
> > Univ. Udine, via delle Scienze 208, 33100 Udine,<br>
> Italy<br>
> > Phone <a href="tel:%2B39-0432-558216" value="+390432558216">+39-0432-558216</a>, fax <a href="tel:%2B39-0432-558222" value="+390432558222">+39-0432-558222</a><br>
> ><br>
> > _______________________________________________<br>
> > Pw_forum mailing list<br>
> > <a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br>
> > <a href="http://pwscf.org/mailman/listinfo/pw_forum" target="_blank">http://pwscf.org/mailman/listinfo/pw_forum</a><br>
> ><br>
> ><br>
> ><br>
> > _______________________________________________<br>
> > Pw_forum mailing list<br>
> > <a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br>
> > <a href="http://pwscf.org/mailman/listinfo/pw_forum" target="_blank">http://pwscf.org/mailman/listinfo/pw_forum</a><br>
><br>
> --<br>
> Paolo Giannozzi, Dept. Chemistry&Physics&Environment,<br>
> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy<br>
> Phone <a href="tel:%2B39-0432-558216" value="+390432558216">+39-0432-558216</a>, fax <a href="tel:%2B39-0432-558222" value="+390432558222">+39-0432-558222</a><br>
><br>
> _______________________________________________<br>
> Pw_forum mailing list<br>
> <a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br>
> <a href="http://pwscf.org/mailman/listinfo/pw_forum" target="_blank">http://pwscf.org/mailman/listinfo/pw_forum</a><br>
><br>
><br>
><br>
> _______________________________________________<br>
> Pw_forum mailing list<br>
> <a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br>
> <a href="http://pwscf.org/mailman/listinfo/pw_forum" target="_blank">http://pwscf.org/mailman/listinfo/pw_forum</a><br>
<br>
--<br>
Paolo Giannozzi, Dept. Chemistry&Physics&Environment,<br>
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy<br>
Phone <a href="tel:%2B39-0432-558216" value="+390432558216">+39-0432-558216</a>, fax <a href="tel:%2B39-0432-558222" value="+390432558222">+39-0432-558222</a><br>
<br>
_______________________________________________<br>
Pw_forum mailing list<br>
<a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br>
<a href="http://pwscf.org/mailman/listinfo/pw_forum" target="_blank">http://pwscf.org/mailman/listinfo/pw_forum</a><br>
</div></div></blockquote></div><br></div>