Error in BSE calculations of hBN-2D

[ManishKumar#1]

Dear All,

While performing the BSE calculations of the tutorials, I am getting the following error:

Command: "yambo -J 2D_WR_WC -F yambo_BSE.in"
Output:


_| _| _|_| _| _| _|_|_| _|_|
_| _| _| _| _|_| _|_| _| _| _| _|
_| _|_|_|_| _| _| _| _|_|_| _| _|
_| _| _| _| _| _| _| _| _|
_| _| _| _| _| _|_|_| _|_|



<---> [01] MPI/OPENMP structure, Files & I/O Directories
<---> MPI Cores-Threads : 1(CPU)-24(threads)
<---> [02] CORE Variables Setup
<---> [02.01] Unit cells
<---> [02.02] Symmetries
<---> [02.03] Reciprocal space
<---> [02.04] K-grid lattice
<---> Grid dimensions : 6 6
<---> [02.05] Energies & Occupations
<---> [03] Transferred momenta grid and indexing
<---> [04] Coloumb potential Random Integration (RIM)
<---> [05] Coloumb potential CutOff box z
<---> [06] External/Internal QP corrections
<---> [QP_apply] Action to be applied: USER defined scissor
<---> [06.01] QP corrections report
<---> [01] Dipoles
<---> [DIP] Checking dipoles header
<---> [02] Dynamic Dielectric Matrix (PPA)
<---> [03] Bethe Salpeter Equation @q1
<---> [03.01] Transition Groups build-up @q1
<---> [BSK] Size (resonant): 216
<---> [BSK] (total): 216
<---> [BSK] Matricies : 1
<---> [03.02] Independent Particles properties @q1
<---> [DIP] Checking dipoles header
<---> [03.03] BSE Kernel @q1 (Resonant CORRRELATION EXCHANGE)
<---> Loading BSE kernel |########################################| [100%] --(E) --(X)
<---> [04] BSE solver(s) @q1
<---> [LA] SERIAL linear algebra
<---> [04.01] Diago Solver @q1
<---> BSK diagonalize | | [000%] --(E) --(X)

[ERROR] STOP signal received while in[04.01] Diago Solver @q1
[ERROR]LINEAR ALGEBRA driver [SERIAL_HERMITIAN_diagonalization]performing CHEEV

What could be the possible cause of this error and what are the possible solutions?

Thanks and Regards,
Manish Kumar
Research Scholar,
IIT Delhi, India

[Daniele Varsano]

Dear Manish,
assuming you are using the very same file of the tutorial, the problem could be in the compilation of the code (links with linear algebra libraries), can you post your config.log file?
In the case instead you modified some of the input files, please post your input and report.
Please use the attachments in the post, you need to rename the file adding a allowed suffix (e.g. .txt, .zip etc.).
Best,
Daniele

[ManishKumar#1]

Dear Daniele,

Thanks for the prompt reply. Please find the attached input and log files that can help in debugging.

Thanks and Regards,
Manish Kumar

[Daniele Varsano]

Dear Manish,
I can see that you are trying to diagonalise a BSE matrix that was calculated in a previous run.
Was that run terminated successfully?
Can you try to delete the files: ./2D_WR_WC//ndb.BS* and rerun the calculation?
If it crashes, please post the new report and log files?
Best,
Daniele

[ManishKumar#1]

Dear Daniele,

The BSE calculation didn't run successfully. Now, I have deleted the file ./2D_WR_WC/ndb.BS*, but the result is same. Please find the attached report file.

Thanks and Regards,
Manish

[Daniele Varsano]

Dear Manish,
please sign your post with your name and affiliation, you can do once for all by filling the signature in your user profile.
There are problem in diagonalizing the excitonic matrix, and it is not easy to understand what is going wrong.

Can you post the results of :

Code: Select all

ncdump  ./2D_WR_WC//ndb.BS_Q1_CPU_0 > bse.txt

if you do not have the ncdump command in your path you can find it in the yambo installation directory:

Code: Select all

yambo/lib/netcdf/netcdf-c-*/ncdump/ncdump

Best,

Daniele

[ManishKumar#1]

Dear Daniele,

Please find the attached file "bse.txt".

Best Regards,
Manish

[Daniele Varsano]

Dear Manish,

as you can see there are many NaNf in the matrix.
Can you do the same for the ndb.dipoles* and ndb.pp* in order try to spot the origin of the problem?

A possibility is that there is a problem in the yambo compilation.
Did you succeed in doing a GW calculation?

Best,
Daniele

[ManishKumar#1]

Dear Daniele,

The GW calculation is running, but not giving the desired output.
I am attaching the files ndb.dipoles* and ndb.pp*. In the ndb.pp_fragment_1, there are entries of NaNf, rest of the files have the floating values.

Best Regards,
Manish

[ManishKumar#1]

Dear Daniele,

Please find the attached config.log file for reference.

Best Regards,
Manish