<div dir="ltr">Hi<div><br></div><div>This maybe the largest hybrid DFT calculation for a 3D periodic system that I have ever heard of.</div><div><br></div><div>You may also check whether you really need a <span style="font-size:12.8px">CUTOFF_RADIUS of 10 A in your PBE0-TC-LRC functional. The density matrix of insulators decays exponentially with SQRT(E_gap)*r12 (see Eqn. 16 of this paper, <a href="http://aip.scitation.org/doi/10.1063/1.2931945">http://aip.scitation.org/doi/10.1063/1.2931945</a>). For system with a sizeable band gap, you can usually get converged electronic properties with a cutoff radius of 6 A or even smaller. I would suggest you to check the convergence of your target properties with respect to </span><span style="font-size:12.8px">CUTOFF_RADIUS in smaller cells, and then choose the </span><span style="font-size:12.8px">CUTOFF_RADIUS that gives the best balance of accuracy and cost for your calculation on the large supercell. </span><span style="font-size:12.8px">A smaller cutoff radius may help you to reduce the memory usage of the HFX calculation.</span></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px">In addition, if you use OT algorithm for SCF, you can usually reduce the </span><span style="font-size:12.8px">MAX_SCF parameter in &SCF subsection to 20 or 30, in order to take advanrage of the preconditioning. This will not reduce the memory usage, but it may help to accelerate the SCF convergence. </span></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px">SL</span></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px"><br></span></div></div><div class="gmail_extra"><br><div class="gmail_quote">On 27 September 2017 at 15:56, Xiaoming Wang <span dir="ltr"><<a href="mailto:wxia...@gmail.com" target="_blank">wxia...@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hi all,<div><br></div><div>Is there any way to reduce the memory usage for HFX calculations? I am dealing with a large scale HFX calculation with the box size of 47*47*47 A and 22560 atoms. After testing, I find that 30Gb mem is required for each MPI task. I am using the MPI/Openmp version. Following is my input, any suggestion on the improvement of the parameters is appreciated. </div><div><br></div><div><div>&GLOBAL</div><div> PROJECT_NAME mysys</div><div> RUN_TYPE GEO_OPT</div><div> PRINT_LEVEL MEDIUM</div><div>&END GLOBAL</div><div>&FORCE_EVAL</div><div> METHOD QS</div><div> &DFT</div><div> CHARGE -1</div><div> LSD</div><div> BASIS_SET_FILE_NAME BASIS_MOLOPT</div><div> BASIS_SET_FILE_NAME BASIS_ADMM_MOLOPT</div><div> POTENTIAL_FILE_NAME GTH_POTENTIALS</div><div> WFN_RESTART_FILE_NAME pbe.wfn</div><div> &QS</div><div> EPS_PGF_ORB 1.0e-32</div><div> &END</div><div> &MGRID</div><div> CUTOFF 250</div><div> REL_CUTOFF 50</div><div> &END MGRID</div><div> &XC</div><div> &XC_FUNCTIONAL</div><div> &PBE</div><div> SCALE_X 0.75</div><div> SCALE_C 1.0</div><div> &END PBE</div><div> &PBE_HOLE_T_C_LR</div><div> CUTOFF_RADIUS 10.0</div><div> SCALE_X 0.25</div><div> &END PBE_HOLE_T_C_LR</div><div> &END XC_FUNCTIONAL</div><div> &HF</div><div> FRACTION 0.25</div><div> &SCREENING</div><div> EPS_SCHWARZ 1.0e-6</div><div> &END SCREENING</div><div> &INTERACTION_POTENTIAL</div><div> POTENTIAL_TYPE TRUNCATED</div><div> CUTOFF_RADIUS 10.0</div><div> T_C_G_DATA ./t_c_g.dat</div><div> &END INTERACTION_POTENTIAL</div><div> &MEMORY</div><div> MAX_MEMORY 32000</div><div> EPS_STORAGE_SCALING 0.1</div><div> &END MEMORY</div><div> &END HF</div><div> &END XC</div><div> &SCF</div><div> MAX_SCF 100</div><div> EPS_SCF 1.0e-6</div><div> CHOLESKY INVERSE</div><div> SCF_GUESS RESTART</div><div> &OT</div><div> ROTATION</div><div> PRECONDITIONER FULL_KINETIC</div><div> ENERGY_GAP 0.001</div><div> &END OT</div><div> &OUTER_SCF</div><div> EPS_SCF 1.0e-6</div><div> MAX_SCF 10</div><div> &END OUTER_SCF</div><div> &END SCF</div><div> &PRINT</div><div> &MULLIKEN OFF</div><div> &END MULLIKEN</div><div> &MO_CUBES</div><div> WRITE_CUBE F</div><div> NHOMO 2</div><div> &END MO_CUBES</div><div> &END PRINT</div><div> &AUXILIARY_DENSITY_MATRIX_<wbr>METHOD</div><div> METHOD BASIS_PROJECTION</div><div> ADMM_PURIFICATION_METHOD MO_DIAG</div><div> &END AUXILIARY_DENSITY_MATRIX_<wbr>METHOD</div><div> &END DFT</div><div> &SUBSYS</div><div> &CELL</div><div> ABC [angstrom] 47.216 47.216 47.216</div><div> ALPHA_BETA_GAMMA [deg] 90 90 90</div><div> PERIODIC XYZ</div><div> SYMMETRY CUBIC</div><div> &END CELL</div><div> &TOPOLOGY</div><div> COORD_FILE_FORMAT XYZ</div><div> COORD_FILE_NAME <a href="http://el.xyz" target="_blank">el.xyz</a></div><div> &END TOPOLOGY</div><div> &KIND Cs</div><div> ELEMENT Cs</div><div> BASIS_SET DZVP-MOLOPT-SR-GTH</div><div> AUX_FIT_BASIS_SET cFIT7</div><div> POTENTIAL GTH-PBE-q9</div><div> &END KIND</div><div> &KIND Pb</div><div> ELEMENT Pb</div><div> BASIS_SET DZVP-MOLOPT-SR-GTH</div><div> AUX_FIT_BASIS_SET cFIT6</div><div> POTENTIAL GTH-PBE-q4</div><div> &END KIND</div><div> &KIND Br</div><div> ELEMENT Br</div><div> BASIS_SET DZVP-MOLOPT-SR-GTH</div><div> AUX_FIT_BASIS_SET cFIT6</div><div> POTENTIAL GTH-PBE-q7</div><div> &END KIND</div><div> &END SUBSYS</div><div>&END FORCE_EVAL</div><div>&MOTION</div><div> &GEO_OPT</div><div> MAX_ITER 500</div><div> OPTIMIZER LBFGS</div><div> &END GEO_OPT</div><div>&END MOTION</div></div><span class="HOEnZb"><font color="#888888"><div><br></div></font></span></div><span class="HOEnZb"><font color="#888888">
<p></p>
-- <br>
You received this message because you are subscribed to the Google Groups "cp2k" group.<br>
To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:cp2k+uns...@googlegroups.com" target="_blank">cp2k+unsubscribe@googlegroups.<wbr>com</a>.<br>
To post to this group, send email to <a href="mailto:cp...@googlegroups.com" target="_blank">cp...@googlegroups.com</a>.<br>
Visit this group at <a href="https://groups.google.com/group/cp2k" target="_blank">https://groups.google.com/<wbr>group/cp2k</a>.<br>
For more options, visit <a href="https://groups.google.com/d/optout" target="_blank">https://groups.google.com/d/<wbr>optout</a>.<br>
</font></span></blockquote></div><br></div>