Dear CP2K developers and users, <br /><div><br /></div><div>I am testing a stress tensor calculation example with my newly compiled CP2K-2025.1 on Platinum 9282 CPUs and I found something wrong:</div><div><br /></div><div>With the binary compiled with "--with-gcc=install --with-intel=no --with-mpich=install<b> --with-mkl=system</b>" tags , the MPI threw an error and the calculation cannot be performed if multiple k-points (say, 2*1*1) were to be sampled, but normally terminated if only gamma point was considered. Since the version compiled with "--with-gcc=
install --with-intel=no --with-mpich=install <b>--with-openblas=install</b>" can perform the calculation with multiple k-points without error (though much solwer than the apptainer image), I suspect that there is something wrong in the version compiled with MKL and wonder if there is any workaround owning to the high performance of MKL.</div><div><br /></div><div>The input file looks like this:</div><div>'''</div><div>&GLOBAL<br /> PROJECT StressTensor<br /> PRINT_LEVEL LOW<br /> RUN_TYPE GEO_OPT<br />&END GLOBAL</div><div><br />&FORCE_EVAL<br /> METHOD Quickstep<br /> STRESS_TENSOR ANALYTICAL<br /> &SUBSYS<br /> &CELL<br /> A <br /> B <br /> C <br /> PERIODIC XYZ #Direction(s) of applied PBC (geometry aspect)<br /> &END CELL<br /> &COORD</div><div><br /></div><div> &END COORD<br /> &KIND Cu<br /> ELEMENT Cu<br /> BASIS_SET DZVP-MOLOPT-SR-GTH-q11<br /> POTENTIAL GTH-PBE<br /> &END KIND<br /> &KIND H<br /> ELEMENT H<br /> BASIS_SET TZVP-MOLOPT-GTH-q1<br /> POTENTIAL GTH-PBE<br /> &END KIND<br /> &KIND O<br /> ELEMENT O<br /> BASIS_SET TZVP-MOLOPT-GTH-q6<br /> POTENTIAL GTH-PBE<br /> &END KIND<br /> &END SUBSYS<br /><br /> &DFT<br /> BASIS_SET_FILE_NAME BASIS_MOLOPT<br /> POTENTIAL_FILE_NAME POTENTIAL<br /> CHARGE 0 #Net charge<br /> MULTIPLICITY 1 #Spin multiplicity<br /> &KPOINTS<br /> SCHEME MONKHORST-PACK 2 1 1<br /> &END KPOINTS<br /> &QS<br /> EPS_DEFAULT 1.0E-12 #Set all EPS_xxx to values such that the energy will be correct up to this value<br /> &END QS<br /> &POISSON<br /> PERIODIC XYZ #Direction(s) of PBC for calculating electrostatics<br /> PSOLVER PERIODIC #The way to solve Poisson equation<br /> &END POISSON<br /> &XC<br /> &XC_FUNCTIONAL PBE<br /> &END XC_FUNCTIONAL<br /> &VDW_POTENTIAL<br /> POTENTIAL_TYPE PAIR_POTENTIAL<br /> &PAIR_POTENTIAL<br /> PARAMETER_FILE_NAME dftd3.dat<br /> TYPE DFTD3(BJ)<br /> REFERENCE_FUNCTIONAL PBE<br /> CALCULATE_C9_TERM T #Calculate C9-related three-body term, more accurate for large system<br /> &END PAIR_POTENTIAL<br /> &END VDW_POTENTIAL<br /> &END XC<br /> &MGRID<br /> CUTOFF 480<br /> REL_CUTOFF 60<br /> NGRIDS 5 #The number of multigrids to use. 5 is optimal for MOLOPT-GTH basis sets<br /> &END MGRID<br /> &SCF<br /> MAX_SCF 256<br /> EPS_SCF 1.0E-06 #Convergence threshold of density matrix of inner SCF<br /> &DIAGONALIZATION<br /> ALGORITHM STANDARD #Algorithm for diagonalization<br /> &END DIAGONALIZATION<br /> &MIXING #How to mix old and new density matrices<br /> METHOD BROYDEN_MIXING #PULAY_MIXING is also a good alternative<br /> ALPHA 0.4 #Default. Mixing 40% of new density matrix with the old one<br /> NBROYDEN 8 #Default is 4. Number of previous steps stored for the actual mixing scheme<br /> &END MIXING<br /> &SMEAR<br /> METHOD FERMI_DIRAC<br /> ELECTRONIC_TEMPERATURE 30 #Electronic temperature of Fermi-Dirac smearing in K<br /> &END SMEAR<br /> ADDED_MOS 504 #Number of virtual MOs to solve<br /> &PRINT<br /> &RESTART #Note: Use "&RESTART OFF" can prevent generating .wfn file<br /> BACKUP_COPIES 0 #Maximum number of backup copies of wfn file. 0 means never<br /> &END RESTART<br /> &END PRINT<br /> &END SCF<br /> &END DFT<br /><br /> &PRINT<br /> &STRESS_TENSOR<br /> &END STRESS_TENSOR<br /> &END PRINT<br />&END FORCE_EVAL<br /><br />&MOTION<br /> &GEO_OPT<br /> TYPE MINIMIZATION #Search for minimum<br /> KEEP_SPACE_GROUP F #If T, then space group will be detected and preserved<br /> OPTIMIZER CG #Can also be BFGS or LBFGS<br /> &CG<br /> &LINE_SEARCH<br /> TYPE 2PNT #Two-point extrapolation, cheap while acceptable. Can also be FIT, GOLD<br /> &END LINE_SEARCH<br /> &END CG<br /> MAX_ITER 0 #Maximum number of geometry optimization<br /> MAX_DR 3E-3 #Maximum geometry change<br /> RMS_DR 1.5E-3 #RMS geometry change<br /> MAX_FORCE 4.5E-4 #Maximum force<br /> RMS_FORCE 3E-4 #RMS force<br /> &END GEO_OPT<br /> &CONSTRAINT<br /> &FIXED_ATOMS #Set atoms to be fixed<br /> COMPONENTS_TO_FIX XYZ #Which fractional components will be fixed, can be X, Y, Z, XY, XZ, YZ, XYZ<br /> LIST 5 6 11 12 17 18 23 24 29 30 35 36 41 42 47 48 53 54 59 60 65 66 71 72 77 78 83 84 89 90 95 96 101 102 107 108 113 114 119 120 125 126 131 132 137 138 143 144 149 150 155 156 161 162 167 168 173 174 179 180 185 186 191 192 197 198 203 204 209 210 215 216 221 222 227 228 233 234 239 240 245 246 251 252 257 258 263 264 269 270 275 276 281 282 287 288 293 294 299 300 305 306 311 312 317 318 323 324 329 330 335 336 341 342 347 348 353 354 359 360 365 366 371 372 377 378 383 384 389 390 395 396 401 402 407 408 413 414 419 420 425 426 431 432<br /> &END FIXED_ATOMS<br /> &FIXED_ATOMS #Set atoms to be fixed<br /> COMPONENTS_TO_FIX Z #Which fractional components will be fixed, can be X, Y, Z, XY, XZ, YZ, XYZ<br /> LIST 721..1008<br /> &END FIXED_ATOMS<br /> &END CONSTRAINT<br /> &PRINT<br /> &TRAJECTORY<br /> FORMAT xyz<br /> &END TRAJECTORY<br /> &RESTART<br /> BACKUP_COPIES 0 #Maximum number of backing up restart file, 0 means never<br /> &END RESTART<br /> &END PRINT<br />&END MOTION</div><div>'''</div><div><br /></div><div>BTW, it's found that the binary compiled with openmpi-5.0.6 shows severe performance problems compared to openmpi-4.1.5 or mpich-4.0.3 in our cluster, and so did the apptainer image, but this is not so urgent :)</div><div><br /></div><div>All your suggestions are greatly appreciated.</div>
<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+unsubscribe@googlegroups.com">cp2k+unsubscribe@googlegroups.com</a>.<br />
To view this discussion visit <a href="https://groups.google.com/d/msgid/cp2k/22e8aac6-f5a6-47b3-80f6-f24356ebea5dn%40googlegroups.com?utm_medium=email&utm_source=footer">https://groups.google.com/d/msgid/cp2k/22e8aac6-f5a6-47b3-80f6-f24356ebea5dn%40googlegroups.com</a>.<br />