[Wien] slurm mpi

Peter Blaha pblaha at theochem.tuwien.ac.at
Tue May 7 17:18:20 CEST 2019


So it seems to works now.

The last messages are probably because you are using   -it   with a 
perfectly converged calculation.
Remove (temporarely) the -it  flag from the runsp_lapw

and remember:   -it may only be faster for surfaces and large cells.


On 5/7/19 4:30 PM, webfinder at ukr.net wrote:
>     Dear Prof. Blaha,
> 
>     I'm using intel mpi 2019.3.199
>     the scalapack and blacs libs are located in the intel
>     compilers_and_libraries_2019.3.199 folder
> 
>     OPTIONS file:
>     current:FOPT:-O1 -FR -mp1 -w -prec_div -pc80 -pad -ip -DINTEL_VML
>     -traceback -assume buffered_io -I$(MKLROOT)/include
>     current:FPOPT:-O1 -FR -mp1 -w -prec_div -pc80 -pad -ip -DINTEL_VML
>     -traceback -assume buffered_io -I$(MKLROOT)/include
>     current:LDFLAGS:$(FOPT) -L$(MKLROOT)/lib/intel64 -lpthread -lm -ldl
>     -liomp5
>     current:DPARALLEL:-DParallel
>     current:R_LIBS:-lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core
>     current:FFTWROOT:/gpfs/home/ser/Install/FFTW/
>     current:FFTW_VERSION:FFTW3
>     current:FFTW_LIB:lib
>     current:FFTW_LIBNAME:fftw3
>     current:LIBXCROOT:/gpfs/home/ser/Install/LIBXC/
>     current:LIBXC_FORTRAN:xcf03
>     current:LIBXC_LIBNAME:xc
>     current:LIBXC_LIBDNAME:lib/
>     current:SCALAPACKROOT:/gpfs/softs/cluster/intel/psxe/2019.3/compilers_and_libraries_2019.3.199/linux/mkl/lib/
>     current:SCALAPACK_LIBNAME:mkl_scalapack_lp64
>     current:BLACSROOT:/gpfs/softs/cluster/intel/psxe/2019.3/compilers_and_libraries_2019.3.199/linux/mkl/lib/
>     current:BLACS_LIBNAME:mkl_blacs_intelmpi_lp64
>     current:ELPAROOT:
>     current:ELPA_VERSION:
>     current:MPIRUN:srun -K -N_nodes_ -n_NP_ -r_offset_ _PINNING_ _EXEC_ 
>         (changed to mpirun)
>     current:CORES_PER_NODE:1
>     current:MKL_TARGET_ARCH:intel64
> 
> 
>     part of .bashrc
> 
>     module load gcc/7.3.0
>     module add compiler/intel/2019.3.199
>     module load mpi/intel/2019.3.199
>     source
>     /gpfs/softs/cluster/intel/psxe/2019.3/compilers_and_libraries_2019.3.199/linux/bin/compilervars.sh
>     intel64
>     source
>     /gpfs/softs/cluster/intel/psxe/2019.3/compilers_and_libraries_2019.3.199/linux/mpi/intel64/bin/mpivars.sh
> 
>     in the interactive mode
> 
>     mpirun -np 4 $WIENROOT/lapw0_mpi lapw0.def
> 
>     results in LAPW0 END
> 
>     Actually, after I commented the following line in my script
>     "if($natom < $nproc) set nproc0=$natom"
>     the "permission denied" error disappeared and mpi starts with the
>     following output:
> 
>     32 nodes for this job: n073 n073 n073 n073 n073 n073 n073 n073 n073
>     n073 n073 n073 n073 n073 n073 n073 n074 n074 n074 n074 n074 n074
>     n074 n074 n074 n074 n074 n074 n074 n074 n074 n074
>       LAPW0 END
>     [1]    Done                          mpirun -n 32 -machinefile
>     .machine0 /gpfs/home/ser/wienroot_v18/lapw0_mpi lapw0.def >> .time00
>     Force-convergence not possible. Forces not present.
>       LAPW1 END
>     [1]  + Done                          ( cd $PWD; $t $ttt; rm -f
>     .lock_$lockfile[$p] ) >> .time1_$loop
>       LAPW1 END
>     [1]  + Done                          ( cd $PWD; $t $ttt; rm -f
>     .lock_$lockfile[$p] ) >> .time1_$loop
>     LAPW2 - FERMI; weights written
>     LAPW2 - FERMI; weights written
>       CORE  END
>       CORE  END
>       MIXER END
> 
>     At the same time in dayfile:
>     Intel MKL ERROR: Parameter 3 was incorrect on entry to DGEMM .
> 
>     Intel MKL ERROR: Parameter 3 was incorrect on entry to DGEMM .
> 
>     Intel MKL ERROR: Parameter 3 was incorrect on entry to DGEMM .
> 
>     Intel MKL ERROR: Parameter 8 was incorrect on entry to DGEMM .
> 
>     Intel MKL ERROR: Parameter 8 was incorrect on entry to DGEMM .
>     ....
> 
>     and in scf file
>     :seclit_par:  estimate of singular value, factor:   0.3125E-01 
>     0.1000E-14
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>         1  0.8207E-20
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>         2  0.6228E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>         3  0.6073E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>         4  0.6256E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>         5  0.9136E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>         6  0.9098E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>         7  0.7724E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>         8  0.7724E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>         9  0.7534E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>        10  0.2265E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>        11  0.2059E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>        12  0.2059E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>        13  0.8401E-18
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>        14  0.8294E-18
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>        15  0.1019E-19
>     :WARN  :seclit_par-stability trick active for: eigenvalue, sproj_ii 
>        16  0.2041E-19
> 
> 
>     This messages are absent in case of k-point parallelization
> 
> 
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> 

-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at/TC_Blaha
--------------------------------------------------------------------------


More information about the Wien mailing list