[Wien] MBJ Problem

[Saeid Jalali]

Dear all,

Recently, I reported the following problem concerning the MBJ potential on
our manipulated carbon-based compound:

http://zeus.theochem.tuwien.ac.at/pipermail/wien/2010-October/013921.html

 

The problem is now fixed. Here, I would/should first report that the problem
was not due to the corrected BRJ subroutine. The BRJ subroutine indeed does
its job properly for low charge density in the interstitial region.

 

Although the source of the problem was not directly related to the WIEN2k
code, it may be useful to be reported here.

The story is as follows:

We installed the v10.1 of the WIEN2k code including the new corrected brj.f
using  l_cprof_p_11.1.072.tgz and its MKL  therein, 10.2.5.035 , on two
PC's-- PC1 under FC10 with 4 cores and 8 GB and PC2 under FC12 with 2 cores
and 2 GB RAM memories. We would run two carbon based cases,case1 and case2 ,
with low charge density in their interstitial regions. The number of
nonequivalent atoms of the case1 is much larger than that of case2. The
number of symmetry operations of case1 is much less than that of case2. The
PC2 could not handle the heavy case1, but the light case2. The heavy case1
could be handled under more powerful PC1.

We performed both of the regular GGA and MBJ calculations for case2 using
PC2 successfully-thanks to Peter and Martin.

We performed the regular GGA for case1 using PC1 successfully. However,
lapw2 problem occurred for the case1 during MBJ calculations. Here, we
reported the problem (see the above link), as we thought that it may be due
to the MBJ subroutine.

We then tried to repeat the successful MBJ calculations of the case2 using
PC1. Surprisingly, the MBJ calculations for the successful case2 stopped
with similar error under PC1. Therefore, at this step we concluded that the
problem may be due to the FC10  of the PC1 compared to the FC12 of the PC2.
We then installed FC12 on another new hard disk under PC1 as well, and
repeated installation of WIEN2k exactly similar to our last installation. We
tested some simple cases under the new hard disk on PC1 with FC12
successfully. We then ran the MBJ for the case1 under new setup of the PC1.
We encountered error in lapw1 (simply error in parallel). We also observed
the speed of calculations are increased by factor 3-three times slower. We
then changed the hard disk. We put the hard disk  of PC2 with its FC12 and
WIEN2k installed on it on PC1. We turned on the PC1 by the hard disk of PC2.
Now PC1 is up under FC12. We ran a simple test. We found segmentation fault
error. We then under the new CPU recompiled once more the WIEN2k code on
PC1. Then we could run successfully simple cases without further attempts. 

 

At this stage we ran our heavy case1 under FC12 on PC1 using the hard disk
of PC2. Surprisingly, it works fine and the case is running smoothly with no
jump in :DIS or :FER.

 

Some strange problems that we cannot explain it.

 

Why FC10 cannot run the MBJ for our case, but can run the regular GGA for
our case and the MBJ for other simple cases?

Why a hard disk with FC12 cannot run MBJ for our case, but can run the MBJ
for other simple cases?

Why changing the hard disk with FC12 can fix the problem?

 

Sincerely yours,

S. Jalali

/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/

Saeid Jalali Asadabadi,

Department of Physics, Faculty of Science,

University of Isfahan (UI), Hezar Gerib Avenue,

81744 Isfahan, Iran.

Phones:

Dep. of Phys.   :+98-0311-793 2435

Office                  :+98-0311-793 4176

Fax No.                :+98-0311-793 2409

E-mail                  :sjalali at phys.ui.ac.ir

                               :sjalali at sci.ui.ac.ir

                               :sjalali at mailaps.org

                                :saeid.jalali.asadabadi at gmail.com

                                :s_jalali_a at yahoo.com

Homepage          :http://sci.ui.ac.ir/~sjalali

www                      :http://www.ui.ac.ir

/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20101029/6f3de284/attachment.htm>