[CP2K-user] [CP2K:20049] How to limit RAM when calculating SP in large supercells

[Krack Matthias]

Hi Eugene

You are running a pretty large system (with respect to cell size and number of atoms) even for a calculation at plain KS-DFT level. You have simply to invest more computational resources, i.e. more compute nodes. Alternatively, you can use a computationally less demanding methodological approach like a tight-binding (TB) method.

HTH

Matthias

From: cp2k at googlegroups.com <cp2k at googlegroups.com> on behalf of Ir Iridium <192.22ir at gmail.com>
Date: Thursday, 21 March 2024 at 10:20
To: cp2k <cp2k at googlegroups.com>
Subject: [CP2K:20049] How to limit RAM when calculating SP in large supercells
Dear CP2K Community,

I am a new user of CP2K and am constantly facing issues with insufficient memory when calculating large supercells. According to the manual, memory can be limited in the HF section. However, during energy calculations, this section is absent from the input file. After starting the calculation, it aborts when the program performs the SCF calculation without any error. This happens when the program takes up all available memory. Therefore, I would like to clarify how to limit the memory usage when calculating the SP of a large supercell.

I am using CP2K version 2024.1. To run the calculation, I use the following commands:

export OMP_NUM_THREADS=1
mpirun -np 26 cp2k.psmp

&GLOBAL
  PROJECT 1ver0-23857_ASC-28
  PRINT_LEVEL LOW
  RUN_TYPE ENERGY
&END GLOBAL

&FORCE_EVAL
  METHOD Quickstep
  &SUBSYS
    &CELL
      A    68.15734400     0.00000000     0.00000000
      B     0.00000000    42.98582200     0.00000000
      C     0.00000000     0.00000000    52.38106900
      PERIODIC XYZ #Direction(s) of applied PBC (geometry aspect)
    &END CELL
    &COORD
    &END COORD
    &KIND Se
      ELEMENT Se
      BASIS_SET DZVP-MOLOPT-SR-GTH-q6
      POTENTIAL GTH-PBE
    &END KIND
    &KIND O
      ELEMENT O
      BASIS_SET DZVP-MOLOPT-SR-GTH-q6
      POTENTIAL GTH-PBE
    &END KIND
    &KIND N
      ELEMENT N
      BASIS_SET DZVP-MOLOPT-SR-GTH-q5
      POTENTIAL GTH-PBE
    &END KIND
    &KIND C
      ELEMENT C
      BASIS_SET DZVP-MOLOPT-SR-GTH-q4
      POTENTIAL GTH-PBE
    &END KIND
    &KIND H
      ELEMENT H
      BASIS_SET DZVP-MOLOPT-SR-GTH-q1
      POTENTIAL GTH-PBE
    &END KIND
  &END SUBSYS

  &DFT
    BASIS_SET_FILE_NAME  BASIS_MOLOPT
    POTENTIAL_FILE_NAME  POTENTIAL
#   WFN_RESTART_FILE_NAME 1ver0-23857_ASC-28-RESTART.wfn
    CHARGE    0 #Net charge
    MULTIPLICITY    1 #Spin multiplicity
    &QS
      EPS_DEFAULT 1.0E-11 #Set all EPS_xxx to values such that the energy will be correct up to this value
    &END QS
    &POISSON
      PERIODIC XYZ #Direction(s) of PBC for calculating electrostatics
      PSOLVER PERIODIC #The way to solve Poisson equation
    &END POISSON
    &XC
      &XC_FUNCTIONAL PBE
      &END XC_FUNCTIONAL
      &VDW_POTENTIAL
        POTENTIAL_TYPE PAIR_POTENTIAL
        &PAIR_POTENTIAL
          PARAMETER_FILE_NAME dftd3.dat
          TYPE DFTD3(BJ)
          REFERENCE_FUNCTIONAL PBE
          #CALCULATE_C9_TERM T #Calculate C9-related three-body term, more accurate for large system
        &END PAIR_POTENTIAL
      &END VDW_POTENTIAL
    &END XC
    &MGRID
      CUTOFF  350
      REL_CUTOFF  50
    &END MGRID
    &SCF
      MAX_SCF 128
      EPS_SCF 5.0E-06 #Convergence threshold of density matrix of inner SCF
#     SCF_GUESS RESTART #Use wavefunction from WFN_RESTART_FILE_NAME file as initial guess
      &DIAGONALIZATION
        ALGORITHM STANDARD #Algorithm for diagonalization
      &END DIAGONALIZATION
      &MIXING #How to mix old and new density matrices
        METHOD BROYDEN_MIXING #PULAY_MIXING is also a good alternative
        ALPHA 0.4 #Default. Mixing 40% of new density matrix with the old one
        NBROYDEN 8 #Default is 4. Number of previous steps stored for the actual mixing scheme
      &END MIXING
      &PRINT
        &RESTART #Note: Use "&RESTART OFF" can prevent generating .wfn file
          BACKUP_COPIES 0 #Maximum number of backup copies of wfn file. 0 means never
        &END RESTART
      &END PRINT
    &END SCF
    &PRINT
      &MO_MOLDEN #Exporting .molden file containing wavefunction information
        NDIGITS 9 #Output orbital coefficients if absolute value is larger than 1E-9
      &END MO_MOLDEN
    &END PRINT
  &END DFT
&END FORCE_EVAL

For better understanding of the problem, I have attached the relevant calculation files. I would be very grateful if you could provide recommendations on limiting memory usage in such cases.

Thank you in advance for your help, and I look forward to your response.

Best regards,
Eugene
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com<mailto:cp2k+unsubscribe at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/cdae1bfe-0490-4be0-b641-e54f60fcb406n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/cdae1bfe-0490-4be0-b641-e54f60fcb406n%40googlegroups.com?utm_medium=email&utm_source=footer>.

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/ZRAP278MB08270BDFA2BD96D298016C1AF4322%40ZRAP278MB0827.CHEP278.PROD.OUTLOOK.COM.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240321/b4f3bcba/attachment-0001.htm>