GW calculation terminating without any error

[Akash]

Dear users,

I am performing GW plasmon pole calculation on a system containing 288 electrons using mpirun with 24 parallel cores. The calculation runs for some time but eventually stops without any error. I am attaching my input files. Please suggest me a solution.

Regards
Akash
SCMS, SavitriBai Phule Pune University, Pune.

[Daniele Varsano]

Dear Akash,

it is possible you are running out of memory. If you post the log file, we can have some more insight.
Anyway, you can try to rerun your calculation by specifying your parallel strategy in a way to distribute the memory:

X_and_IO_CPU= "1 1 1 4 6" # [PARALLEL] CPUs for each role
X_and_IO_ROLEs= "q g k c v" # [PARALLEL] CPUs roles (q,g,k,c,v)
DIP_CPU= "1 4 6" # [PARALLEL] CPUs for each role
DIP_ROLEs= "k c v"

Besides that, please note:

Code: Select all

NGsBlkXp= 1 RL 

You are using just one RL in the response function, this is way out of convergence, you will need to check the convergence of this variable, you can have a look at the tutorials on GW that are in the yambo wiki page.

Code: Select all

%QPkrange                        # [GW] QP generalized Kpoint/Band indices
1|7|1|154|
%

You are trying to correct 1078 QP energies, it is a very large number, you have 96 occupied bands, are you sure you are interested in the deep bands.
Usually one it is interested in few bands around the gap. Calculating so many bands, it is a waste of time and resources.

Best,
Daniele

[Akash]

Dear Daniele,
Thank you for taking out your precious time and replying to my query. Your suggestion worked. It was the memory issue. I have increased the memory and the code is running smoothly.

Regards
Akash