BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES

Deals with issues related to computation of optical spectra in reciprocal space: RPA, TDDFT, local field effects.

Moderators: Davide Sangalli, andrea.ferretti, myrta gruning, andrea marini, Daniele Varsano, Conor Hogan

Post Reply
joydev.asadtala
Posts: 8
Joined: Tue Dec 08, 2020 6:35 am

BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES

Post by joydev.asadtala » Fri Dec 18, 2020 5:16 am

Dear developer,

I am doing the optics calculation using ALDA kernel.
For
FxcGRLc= 10 RL # [TDDFT] XC-kernel RL size
NGsBlkXd= 10 RL # [Xd] Response block size

the calculation exit with out any error. But when i increasing the value of FxcGRLc & NGsBlkXd, I am getting the below error message.

===================================================================================
= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
= PID 195578 RUNNING AT n41
= EXIT CODE: 9
= CLEANING UP REMAINING PROCESSES
= YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
===================================================================================
Intel(R) MPI Library troubleshooting guide:
https://software.intel.com/node/561764
===================================================================================

Is this for memory problem issue?

I am attaching all the input files.

Best,
Joydev De
PhD student, NISER, India
You do not have the required permissions to view the files attached to this post.

User avatar
Daniele Varsano
Posts: 3816
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES

Post by Daniele Varsano » Fri Dec 18, 2020 8:50 am

Dear Joydev,
it is possible it is a memory problem, but it seems also from the LOG that yambo wants to consider much more than 100RL.

Code: Select all

[PARALLEL distribution for RL vectors(X) on 4 CPU] Loaded/Total (Percentual):1605/11449(14%)
I do not know what it is happening.
You can try to assign the dimension of the matrix in Ry, e.g.

Code: Select all

FxcGRLc= 1            Ry      # [TDDFT] XC-kernel RL size
NGsBlkXd= 1            Ry      # [Xd] Response block size
than moving to larger values and see if it runs correctly.

Other option, is to solve the TDDFT equation in transition space:

Code: Select all

yambo -o b -k alda -y h  
In this way, you do not need to specify the dimension of the kernel as all G vectors are automatically taken into consideration and you do not need to solve the Dyson equation for all the frequencies.

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/

Post Reply