GW computation stuck for a large system

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
DavidPolito93
Posts: 71
Joined: Sat Jun 06, 2020 10:43 am

GW computation stuck for a large system

Post by DavidPolito93 » Fri Mar 25, 2022 1:08 pm

Dear All,

I am trying to perform GW computations for a molecule adsorbed on a bilayer material. The system is sufficiently large to be treated like a finite system (i.e. k-sampling 1 1 1).

I have however a problem, my computation gets stuck at the step which computes the Chi and then crushes. It is probably related to a memory issue I think.

I am attaching the inputs and the logs files. I am using Yambo 5.0.4

Do you have any suggestion?

Best,

Davide Romanin
You do not have the required permissions to view the files attached to this post.
-----------------------------------------------------
Assistant Professor
Polytech - Paris-Saclay University
C2N, CNRS
10 Bd Thomas Gobert
91120 Palaiseau

User avatar
Daniele Varsano
Posts: 3773
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: GW computation stuck for a large system

Post by Daniele Varsano » Fri Mar 25, 2022 2:12 pm

Dar Davide,

it seems the problem arise in the linear algebra to obtain the X, you can try to recompile the code using parallel linear algebra (SCALAPACK) (--enable-par-linalg ).

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/

DavidPolito93
Posts: 71
Joined: Sat Jun 06, 2020 10:43 am

Re: GW computation stuck for a large system

Post by DavidPolito93 » Fri Mar 25, 2022 7:50 pm

Dear Daniele,

I did it like you suggested:

make clean

./configure --enable-par-linalg

Then I recompiled and no errors appeared.

Anyway, I still have this:

<14m-25s> P1-r4i6n29: [LA] SERIAL linear algebra


Si I assume it didn't go through?

Davide
-----------------------------------------------------
Assistant Professor
Polytech - Paris-Saclay University
C2N, CNRS
10 Bd Thomas Gobert
91120 Palaiseau

User avatar
Daniele Varsano
Posts: 3773
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: GW computation stuck for a large system

Post by Daniele Varsano » Tue Mar 29, 2022 9:22 pm

Dear Davide,
you should inspect your config.log file and see if the scalapack libraries are correctly linked. This is also reported in the config/setup file or in the header of your report file (indicated as SLK).

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