[Wien] overflow on array size calculation

[Rajneesh Chaurasiya]

Dear Wien2k users,

I have upgraded wien2k version from 13 to 16 successfully but when i do the
A2BBO6 (double perovskite) calculation then during the execution of scf
error looks like

forrtl: severe (179): Cannot allocate array - overflow on array size
calculation.
Image  PCRoutineLine
Source
lapw2  0051BD2A  Unknown   Unknown  Unknown
lapw2  0051A8A5  Unknown   Unknown  Unknown
lapw2  004C93C6  Unknown   Unknown  Unknown
lapw2  0047BB96  Unknown   Unknown  Unknown
lapw2  004AD22A  Unknown   Unknown  Unknown
lapw2  00431A4F  fourir_   112
fourir_tmp_.F
lapw2  0045996C  MAIN__747
lapw2_tmp_.F
lapw2  004040AC  Unknown   Unknown  Unknown
libc.so.6  003B5641D9F4  Unknown   Unknown  Unknown
lapw2  00403FB9  Unknown   Unknown  Unknown


While i used maximum 20 nodes, each nodes has 12 processor. So i think
there is no issue of memory and also tried complex calculation but similar
error occurred.
RMT value of each atoms is looking fine.

I have run same calculation in the past using 13 version there was no error
but in wien2k 16 it gives error.


So can you suggest me where will be a possible error.


-- 
Thanks & Regards
Rajneesh Chaurasiya
Research Scholar
IIT Jodhpur, India
Mob. No. +91-9584499697
  +91-7610950803
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] overflow on array size calculation

[Dr. K. C. Bhamu]

Hello Rajneesh,

Check you bashrc file and see whether you defined:

ulimit -s unlimited

if you do not find it copy and past at the bottom of bashrc file.




On Sat, May 20, 2017 at 5:13 PM, Rajneesh Chaurasiya 
wrote:

> Dear Wien2k users,
>
> I have upgraded wien2k version from 13 to 16 successfully but when i do
> the A2BBO6 (double perovskite) calculation then during the execution of scf
> error looks like
>
> forrtl: severe (179): Cannot allocate array - overflow on array size
> calculation.
> Image  PCRoutineLine
> Source
> lapw2  0051BD2A  Unknown   Unknown  Unknown
> lapw2  0051A8A5  Unknown   Unknown  Unknown
> lapw2  004C93C6  Unknown   Unknown  Unknown
> lapw2  0047BB96  Unknown   Unknown  Unknown
> lapw2  004AD22A  Unknown   Unknown  Unknown
> lapw2  00431A4F  fourir_   112
> fourir_tmp_.F
> lapw2  0045996C  MAIN__747
> lapw2_tmp_.F
> lapw2  004040AC  Unknown   Unknown  Unknown
> libc.so.6  003B5641D9F4  Unknown   Unknown  Unknown
> lapw2  00403FB9  Unknown   Unknown  Unknown
>
>
> While i used maximum 20 nodes, each nodes has 12 processor. So i think
> there is no issue of memory and also tried complex calculation but similar
> error occurred.
> RMT value of each atoms is looking fine.
>
> I have run same calculation in the past using 13 version there was no
> error but in wien2k 16 it gives error.
>
>
> So can you suggest me where will be a possible error.
>
>
> --
> Thanks & Regards
> Rajneesh Chaurasiya
> Research Scholar
> IIT Jodhpur, India
> Mob. No. +91-9584499697
>   +91-7610950803
>
> ___
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/
> wien@zeus.theochem.tuwien.ac.at/index.html
>
>
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] overflow on array size calculation

[Laurence Marks]

Sorry, but it is almost certainly nothing to do with ulimit since that
is done in software now (has been for some years).

It is hard to say without other information what is going on. Is this
a large calculation? What does "grep ifft1 *.output2* " give you? Is
this mpi or just k-point parallel?

If it is a large calculation it may be that you need to use
lapw2_vector_split in your .machines file. In some cases lapw2 can run
out of memory.

It could also be that lapw1 crashed, so check that. In principle a
crash of lapw1 should stop the program but sometimes it does not.

On Sat, May 20, 2017 at 7:54 AM, Dr. K. C. Bhamu  wrote:
> Hello Rajneesh,
>
> Check you bashrc file and see whether you defined:
>
> ulimit -s unlimited
>
> if you do not find it copy and past at the bottom of bashrc file.
>
>
>
>
> On Sat, May 20, 2017 at 5:13 PM, Rajneesh Chaurasiya 
> wrote:
>>
>> Dear Wien2k users,
>>
>> I have upgraded wien2k version from 13 to 16 successfully but when i do
>> the A2BBO6 (double perovskite) calculation then during the execution of scf
>> error looks like
>>
>> forrtl: severe (179): Cannot allocate array - overflow on array size
>> calculation.
>> Image  PCRoutineLineSource
>> lapw2  0051BD2A  Unknown   Unknown
>> Unknown
>> lapw2  0051A8A5  Unknown   Unknown
>> Unknown
>> lapw2  004C93C6  Unknown   Unknown
>> Unknown
>> lapw2  0047BB96  Unknown   Unknown
>> Unknown
>> lapw2  004AD22A  Unknown   Unknown
>> Unknown
>> lapw2  00431A4F  fourir_   112
>> fourir_tmp_.F
>> lapw2  0045996C  MAIN__747
>> lapw2_tmp_.F
>> lapw2  004040AC  Unknown   Unknown
>> Unknown
>> libc.so.6  003B5641D9F4  Unknown   Unknown
>> Unknown
>> lapw2  00403FB9  Unknown   Unknown
>> Unknown
>>
>>
>> While i used maximum 20 nodes, each nodes has 12 processor. So i think
>> there is no issue of memory and also tried complex calculation but similar
>> error occurred.
>> RMT value of each atoms is looking fine.
>>
>> I have run same calculation in the past using 13 version there was no
>> error but in wien2k 16 it gives error.
>>
>>
>> So can you suggest me where will be a possible error.
>>
>>
>> --
>> Thanks & Regards
>> Rajneesh Chaurasiya
>> Research Scholar
>> IIT Jodhpur, India
>> Mob. No. +91-9584499697
>>   +91-7610950803
>>
>> ___
>> Wien mailing list
>> Wien@zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>> SEARCH the MAILING-LIST at:
>> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>>
>



-- 
Professor Laurence Marks
"Research is to see what everybody else has seen, and to think what
nobody else has thought", Albert Szent-Gyorgi
www.numis.northwestern.edu ; Corrosion in 4D: MURI4D.numis.northwestern.edu
Partner of the CFW 100% program for gender equity, www.cfw.org/100-percent
Co-Editor, Acta Cryst A
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

[Wien] Optical properties of half-metals

[Rashid Ahmad]

Dear All, 
   Is it possible to calculate optical properties of 
spin-up and spin-down electrons half-metals such as Heuslers, separately using 
WIENk? For example,  refractive index of spin-up channel and spin-down channel 
plotted separately?
 Rashid___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

[Wien] Import error, no module named numpy

[Jameela Fatheema]

 Dear all,
I am running wien2kversion 14 on ubuntu 16.04
The purpose of mycalculations is to get born effective charges for AlN.
Thescf has been run successfully for 1000 kpoints. And I am trying torun the 
command “berrypi -k 6:6:6” but I am getting the followingerror.
ubuntu@ubuntu-7G-Series:~/project/AlNitride/12may$berrypi -k 6:6:6Traceback 
(mostrecent call last): 
File"/home/ubuntu/WIEN2k_14/SRC_BerryPI/BerryPI/berrypi", line20, in  
import numpy #math libraryImportError: Nomodule named numpy
I have tried to makesure that python , numpy , berrypi and wannier90 have been 
installedcorrectly but the error remains.
I would like to knowhow to solve this problem. Kindly if anyone can help.
Thank you
Regards
Jameela FatheemaDepartment of Physics
University of Peshawar
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

[Wien] How to fit birch Murnaghan EOS for the orthorhombic case

[BUSHRA SABIR]

Dear Wien2k users,
For an orthorhombic optimization case I have two queries:
I found two processes to optimize the orthorhombic structure 1. By Prof. Marks  
2. By Dr. Delamora
To optimize the structure file according to the method suggested by Prof. 
Marks, I followed the below steps:
1. Vol optimization with option 1 in "x optimize" script. It will give 
optimized volume.2. Using previously optimized structure, I will do here c/a 
optimization with option 3 from "x optimize". It will give us scaling for c/a. 
Updated c/a in the structure file obtained from this step and3.  Using 
previously optimized structure, I will do here b/a A  optimization with option 
4 from "x optimize". It will give us scaling for b/a. Used optimized b/a from 
here in the structure file obtained from this step after a fit.
Now my first Query is: 
A. Is the final structure file (from step 3) is the well optimised for 
orthorhombic case?B. Is this correct that the last step will give me a  fitted 
birch Murnaghan EOS? 

For the process mentioned in "x optimize" with option 6 bu Delamora:
In the option 6 you can put one of the mentioned numbers
10, 27, 64,125 in which the parameters are varied by 3x3x3 (=27 different 
cases)  'a' is varied by -3 0 3, the same for 'b and 'c', then you have to fit 
a curve to these results

My second query is:Is the fitting process is same for option 6 as for option 2 
or 3 or we need to follow different approach?
Your help will definetly help us to shortout the problem.
Looking forward to hearing from you.
Regards ___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] How to fit birch Murnaghan EOS for the orthorhombic case

[delamora]

In the first case you have to do several calculations, in the option 6 you have 
3x3x3 mesh and if you fit a 3D 2nd degree equation you get the answer with one, 
although long, calculation

f=a+bx+cy+dz+exy...

and the minimum will get you the optimized cell parapeters

Is this the most exact answer? I do not know, because you are fitting a parabola


Saludos


Pablo



De: Wien  en nombre de BUSHRA SABIR 

Enviado: sábado, 20 de mayo de 2017 03:52:53 p. m.
Para: wien@zeus.theochem.tuwien.ac.at
Asunto: [Wien] How to fit birch Murnaghan EOS for the orthorhombic case

Dear Wien2k users,

For an orthorhombic optimization case I have two queries:

I found two processes to optimize the orthorhombic structure 1. By Prof. 
Marks
  2. By Dr. 
Delamora

To optimize the structure file according to the method suggested by Prof. 
Marks, I followed the below steps:

1. Vol optimization with option 1 in "x optimize" script. It will give 
optimized volume.
2. Using previously optimized structure, I will do here c/a optimization with 
option 3 from "x optimize". It will give us scaling for c/a. Updated c/a in the 
structure file obtained from this step and
3.  Using previously optimized structure, I will do here b/a A  optimization 
with option 4 from "x optimize". It will give us scaling for b/a. Used 
optimized b/a from here in the structure file obtained from this step after a 
fit.

Now my first Query is:
A. Is the final structure file (from step 3) is the well optimised for 
orthorhombic case?
B. Is this correct that the last step will give me a  fitted birch Murnaghan 
EOS?

For the process mentioned in "x optimize" with option 6 bu 
Delamora:


In the option 6 you can put one of the mentioned numbers
10, 27, 64,125 in which the parameters are varied by 3x3x3 (=27 different 
cases)  'a' is varied by -3 0 3, the same for 'b and 'c',

 then you have to fit a curve to these results


My second query is:

Is the fitting process is same for option 6 as for option 2 or 3 or we need to 
follow different approach?


Your help will definetly help us to shortout the problem.


Looking forward to hearing from you.


Regards


[http://mail.yimg.com/us.yimg.com/i/mesg/tsmileys2/01.gif]

___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html