B3LYP/D3 Functional Failure/Stalling in LCAO and PW Runs #6695
Description
Details
Hello,
I am writing to report an issue encountered when using the B3LYP functional in a simulation with the LCAO basis set and HYB_GGA_XC_B3LYP + D3 functional.
I was using the v3.9.0.17 Abacus version, compiled by GNU (using the provided toolchain). However, the simulation took an anomalously long amount of time (statistics shown below) and appears to have missed the electron correlation calculation. I am just wondering if this is due to any settings?
TIME STATISTICS
----------------------------------------------------------------
CLASS_NAME NAME TIME/s CALLS AVG/s PER/%
----------------------------------------------------------------
total 80885.87 13 6221.99 100.00
Driver atomic_world 80885.87 1 80885.87 100.00
Relax_Driver relax_driver 80847.23 1 80847.23 99.95
ESolver_KS runner 80847.23 1 80847.23 99.95
HSolverLCAO solve 80397.64 15 5359.84 99.40
HamiltLCAO updateHk 4650.09 180 25.83 5.75
OperatorLCAO init 4586.73 540 8.49 5.67
Veff contributeHR 4413.79 15 294.25 5.46
Gint cal_gint_vl 4413.57 15 294.24 5.46
HSolverLCAO hamiltSolvePsiK 56473.92 180 313.74 69.82
DiagoElpa elpa_solve 56451.56 180 313.62 69.79
elecstate cal_dm 15546.09 15 1036.41 19.22
psiMulPsiMpi pdgemm 15545.40 180 86.36 19.22
LCAO_domain dm2rho 3715.17 15 247.68 4.59
Gint cal_gint_rho 3714.67 15 247.64 4.59
----------------------------------------------------------------
The simulation of a similar system usually yields statistics like this:
My input parameters for the LCAO run were as follows:
INPUT_PARAMETERS
calculation scf
symmetry 1
ecutwfc 100
scf_thr 1e-07
scf_nmax 100
smearing_method gauss
smearing_sigma 0.015
mixing_type broyden
mixing_beta 0.8
basis_type lcao
ks_solver genelpa
kspacing 0.14
dft_functional HYB_GGA_XC_B3LYP
vdw_method d3_0
Additionally, when I tried the same simulation with the PW basis set, the calculation seemed to get stuck at the SCF initiation step. It remained in this state for more than 22 hours, so I am wondering if this might also be related to the settings. Here's the time statistics:
---------------------------------------------------------
DONE(0.747951 SEC) : INIT PLANEWAVE
DONE(20.4249 SEC) : LOCAL POTENTIAL
DONE(21.525 SEC) : NON-LOCAL POTENTIAL
MEMORY FOR PSI (MB) : 3082.53
DONE(21.6953 SEC) : INIT BASIS
================================================================
SELF-CONSISTENT:
================================================================
START CHARGE : atomic
DONE(79.2032 SEC) : INIT SCF
Here is the input file for the PW run:
INPUT_PARAMETERS
calculation scf
symmetry 1
ecutwfc 90.0
scf_thr 1e-08
scf_nmax 100
smearing_method gauss
smearing_sigma 0.015
mixing_type broyden
mixing_beta 0.8
basis_type pw
ks_solver dav_subspace
kspacing 0.14
vdw_method d3_0
dft_functional HYB_GGA_XC_B3LYP
pw_diag_ndim 2
pw_diag_nmax 20
Have you read FAQ on the online manual http://abacus.deepmodeling.com/en/latest/community/faq.html
- Yes, I have read the FAQ part on online manual.
Task list for Issue attackers (only for developers)
- Understand the problem or question described by the user.
- Check if the issue is a known problem or has been addressed in the documentation.
- Test the issue or problem on a similar system or environment, if possible.
- Identify the root cause or provide clarification on the user's question.
- Provide a step-by-step guide, including any necessary resources, to resolve the issue or answer the question.
- If the issue is related to documentation, update the documentation to prevent future confusion (optional).
- If the issue is related to code, consider implementing a fix or improvement (optional).
- Review and incorporate any relevant feedback from users or developers.
- Ensure the user's issue is resolved or their question is answered and close the ticket.