Page 1 of 1

Dynamic Dielectric Matrix [05] Free DIP_iR stops code

Posted: Wed Jun 17, 2015 12:06 pm
by jmullen

I am running an MoS2 system in order work my way through GW and BSE calculations (see viewtopic.php?f=14&t=1019 ). I am increasing the number of k points with which the calculation is performed (9x9, 18x18, 27x27...) in order to get accurate band structure values for the system. Recently, I ran a 54x54 grid and began the convergence process on the same variables under the configuration conditions and the system stops running. Where it stops running is at

[05] Dynamic Dielectric Matrix (PPA)
Alloc DIP_iR
Free DIP_iR

I watch the system memory on the nodes running the calculation and there is no undue stress on the either the RAM, CPU or SWAP at this point in the process and the program does not write any information on stop. Now, I know that anyone here probably needs more information about the system to help with the crash, but my question is more about what is happening in the code when it crashes.

When I look at all other output files (l_em1d_ppa_HF_and_locXC_gw0_##) for all of my other runs, there is no mention of DIP_iR. A novice reading of this seems to suggest that increased MP grid triggers a different routine and that routine is causing me problems. What is DIP_iR and why have I not encountered it in my previous output files?

Additional information:
total number of k-points in run: 1460

Input file:

HF_and_locXC # [R XX] Hartree-Fock Self-energy and Vxc
ppa # [R Xp] Plasmon Pole Approximation
gw0 # [R GW] GoWo Quasiparticle energy levels
em1d # [R Xd] Dynamical Inverse Dielectric Matrix
EXXRLvcs=100 RL # [XX] Exchange RL components
Chimod= "Hartree" # [X] IP/Hartree/ALDA/LRC/BSfxc
% BndsRnXp
1 | 400 | # [Xp] Polarization function bands
NGsBlkXp= 25 RL # [Xp] Response block size
% LongDrXp
1.000000 | 0.000000 | 0.000000 | # [Xp] [cc] Electric Field
PPAPntXp= 27.21138 eV # [Xp] PPA imaginary energy
% GbndRnge
1 | 400 | # [GW] G[W] bands range
GDamping= 0.10000 eV # [GW] G[W] damping
dScStep= 0.10000 eV # [GW] Energy step to evalute Z factors
DysSolver= "n" # [GW] Dyson Equation solver (`n`,`s`,`g`)
%QPkrange # [GW] QP generalized Kpoint/Band indices
623|623| 9|10|
965|965| 9|10|

Re: Dynamic Dielectric Matrix [05] Free DIP_iR stops code

Posted: Mon Jun 22, 2015 1:00 pm
by Davide Sangalli
Dear Jeff,
the reason why the messages Alloc / Free do not appear in smaller runs may be due to the fact that they are written only when the size of the allocated object (DIP_iR in this case) is above a given threshold.

In any case DIP_iR contains "i" times "the expectation value of the r operator" and is a matrix in the valence-conduction space. It is needed to compute the q=0 term of the dielectric matrix.

Can you post the complete log file and report from your run.

Kind regards,

Re: Dynamic Dielectric Matrix [05] Free DIP_iR stops code

Posted: Wed Jul 01, 2015 5:02 pm
by jmullen
Hello Davide,

Thanks for the reply. It turns out that the problem I was having was due to a unrelated file system problem that the admin team was fighting at the same time. This particular issue seems to be resolved