Dear CP2K users, <div><br></div><div>I would like benchmark a small periodic system (11 A x 11 A x 11A) using HSE06 functional with results obtained from VASP,</div><div>Here is my input for DFT section:</div><div><br></div><div><div> &DFT</div><div> BASIS_SET_FILE_NAME BASIS_MOLOPT_UCL</div><div> BASIS_SET_FILE_NAME BASIS_MOLOPT</div><div> BASIS_SET_FILE_NAME BASIS_ADMM_MOLOPT</div><div> BASIS_SET_FILE_NAME BASIS_ADMM</div><div> POTENTIAL_FILE_NAME GTH_POTENTIALS</div><div> WFN_RESTART_FILE_NAME cp2k-RESTART.wfn</div><div> &MGRID</div><div> CUTOFF 320</div><div> COMMENSURATE</div><div> &END MGRID</div><div> &QS</div><div> EXTRAPOLATION PS</div><div> EXTRAPOLATION_ORDER 3</div><div> EPS_DEFAULT 1.0E-11</div><div> EPS_PGF_ORB 1.0E-14</div><div> MAP_CONSISTENT T</div><div> &END QS</div><div> &SCF</div><div> SCF_GUESS RESTART</div><div> EPS_SCF 1.0E-7</div><div> MAX_SCF 300</div><div> ADDED_MOS 100</div><div> &DIAGONALIZATION</div><div> ALGORITHM STANDARD</div><div> &END DIAGONALIZATION</div><div> &SMEAR ON</div><div> METHOD FERMI_DIRAC</div><div> ELECTRONIC_TEMPERATURE [K] 300</div><div> &END SMEAR</div><div> &MIXING</div><div> METHOD BROYDEN_MIXING</div><div> ALPHA 0.2</div><div> BETA 1.5</div><div> NBROYDEN 8</div><div> &END MIXING</div><div> &END SCF</div><div> !&XC</div><div> ! &XC_FUNCTIONAL PBE</div><div> ! &END XC_FUNCTIONAL</div><div> !&END XC</div><div> &XC</div><div> &XC_FUNCTIONAL</div><div> &PBE</div><div> SCALE_X 0.0</div><div> SCALE_C 1.0</div><div> &END PBE</div><div> &XWPBE</div><div> SCALE_X -0.25</div><div> SCALE_X0 1.0</div><div> OMEGA 0.11</div><div> &END XWPBE</div><div> &END XC_FUNCTIONAL</div><div> &HF</div><div> &SCREENING</div><div> EPS_SCHWARZ 1.0E-6</div><div> SCREEN_ON_INITIAL_P T</div><div> &END SCREENING</div><div> &INTERACTION_POTENTIAL</div><div> POTENTIAL_TYPE SHORTRANGE</div><div> OMEGA 0.11</div><div> &END INTERACTION_POTENTIAL</div><div> &MEMORY</div><div> MAX_MEMORY 4000</div><div> EPS_STORAGE_SCALING 0.1</div><div> &END MEMORY</div><div> FRACTION 0.25</div><div> &END HF</div><div> &END XC</div><div> &AUXILIARY_DENSITY_MATRIX_METHOD</div><div> METHOD BASIS_PROJECTION</div><div> ADMM_PURIFICATION_METHOD NONE</div><div> &END AUXILIARY_DENSITY_MATRIX_METHOD</div><div> &PRINT</div><div> &PDOS</div><div> FILENAME pdos</div><div> # print all projected DOS available:</div><div> NLUMO -1</div><div> # split the density by quantum number:</div><div> COMPONENTS</div><div> &END</div><div> &END PRINT</div><div> &END DFT</div></div><div><br></div><div>The calculation restarted from a converged PBE wavefunction.</div><div>However, I found that the calculation is quite "slow" ( Vasp needs 240 seconds for a SCF step, but CP2K needs almost 2400 seconds. Both of them are carried out using a computing node with 24 cores and 120 G memory in total). I understand it is not easy to compare the different software because of very different setups, but I wound expect the ADMM method in CP2K should be much faster.</div><div><br></div><div>Below is the output.</div><div><br></div><div> SCF WAVEFUNCTION OPTIMIZATION<br></div><div><div><br></div><div> Step Update method Time Convergence Total energy Change</div><div> ------------------------------------------------------------------------------</div><div><br></div><div> HFX_MEM_INFO| Est. max. program size before HFX [MiB]: 792</div><div><br></div><div> *** WARNING in hfx_types.F:1287 :: Periodic Hartree Fock calculation ***</div><div> *** requested with use of a truncated or shortrange potential. The cutoff ***</div><div> *** radius is larger than half the minimal cell dimension. This may lead ***</div><div> *** to unphysical total energies. Reduce the cutoff radius in order to ***</div><div> *** avoid possible problems. ***</div><div><br></div><div> HFX_MEM_INFO| Number of cart. primitive ERI's calculated: 11992558561508</div><div> HFX_MEM_INFO| Number of sph. ERI's calculated: 157558545566</div><div> HFX_MEM_INFO| Number of sph. ERI's stored in-core: 16901607068</div><div> HFX_MEM_INFO| Number of sph. ERI's stored on disk: 0</div><div> HFX_MEM_INFO| Number of sph. ERI's calculated on the fly: 91978901962</div><div> HFX_MEM_INFO| Total memory consumption ERI's RAM [MiB]: 13711</div><div> HFX_MEM_INFO| Whereof max-vals [MiB]: 454</div><div> HFX_MEM_INFO| Total compression factor ERI's RAM: 9.41</div><div> HFX_MEM_INFO| Total memory consumption ERI's disk [MiB]: 0</div><div> HFX_MEM_INFO| Total compression factor ERI's disk: 0.00</div><div> HFX_MEM_INFO| Size of density/Fock matrix [MiB]: 24</div><div> HFX_MEM_INFO| Size of buffers [MiB]: 90</div><div> HFX_MEM_INFO| Number of periodic image cells considered: 123</div><div> HFX_MEM_INFO| Est. max. program size after HFX [MiB]: 3582</div><div><br></div><div> 1 NoMix/Diag. 0.20E+00 6553.7 0.12989389 -3154.6382899197 -3.15E+03</div><div><br></div><div> *** WARNING in hfx_types.F:1287 :: Periodic Hartree Fock calculation ***</div><div> *** requested with use of a truncated or shortrange potential. The cutoff ***</div><div> *** radius is larger than half the minimal cell dimension. This may lead ***</div><div> *** to unphysical total energies. Reduce the cutoff radius in order to ***</div><div> *** avoid possible problems. ***</div><div><br></div><div> 2 Broy./Diag. 0.20E+00 2486.1 0.00624233 -3159.6346919624 -5.00E+00</div><div><br></div><div> *** WARNING in hfx_types.F:1287 :: Periodic Hartree Fock calculation ***</div><div> *** requested with use of a truncated or shortrange potential. The cutoff ***</div></div><div><br></div><div><div><br></div></div><div>Is there anything wrong with my input that slows down the calculation?</div><div>In particular, the " ERI's calculated on the fly" is not zero which seems not good according to a slide from "<a href="https://mattatlincoln.github.io/talks/GhentWorkshop/?print-pdf#/">https://mattatlincoln.github.io/talks/GhentWorkshop/?print-pdf#/</a>"</div><div><br></div><div>Thank you very much in advance<br>Best Regards,</div><div>Geng</div><div><br></div>