====== Magnus cluster ====== ===== Info ===== Workload manager - [[https://wiki.storion.ru/linux/slurm|SLURM]] [[http://cest-cms-wiki.skoltech.ru/doku.php?id=start|Wiki for users]] ===== Run script example ===== ===VASP === #!/bin/bash #SBATCH -J my_job_LiMnO2 #name of job #SBATCH -N 1 #number of nodes #SBATCH -n 16 #number of cores #SBATCH -o %x.e%j #out file path #SBATCH -e %x.e%j.err #err file path #SBATCH --mem-per-cpu=7675 #memory limit #SBATCH -p AMG-medium #type of schedule: AMG (24 h limit), AMG-medium (72 h), AMG-long (336 h) module load Compiler/Intel/17u8 module load Q-Ch/VASP/5.4.4_OPT ScriptLang/python/3.6i_2018u3 ulimit -s unlimited # cd /home/user/project_folder # provide the folder if the script is submitted not from project folder mpirun vasp_std >log ===VASP with Occupation Matrix Control=== The same as previous. Differs only by another loaded Q-Ch/VASP/5.4.4_OMC module #!/bin/bash #SBATCH -J my_job_LiMnO2 #name of job #SBATCH -N 1 #number of nodes #SBATCH -n 16 #number of cores #SBATCH -o %x.e%j #out file path #SBATCH -e %x.e%j.err #err file path #SBATCH --mem-per-cpu=7675 #memory limit #SBATCH -p AMG-medium #type of schedule: AMG (24 h limit), AMG-medium (72 h), AMG-long (336 h) module load Compiler/Intel/17u8 Q-Ch/VASP/5.4.4_OMC ScriptLang/python/3.6i_2018u3 ulimit -s unlimited # cd /home/user/project_folder # provide the folder if the script is submitted not from project folder mpirun vasp_std >log === LAMMPS === #!/bin/bash #SBATCH -J meltcool10Kpps # job name, should coincide with input filename #SBATCH -N 1 #SBATCH -n 8 # number of cores #SBATCH -o %x.e%j #SBATCH -e %x.e%j.err #SBATCH --mem-per-cpu=8gb export OMP_NUM_THREADS="'${SLURM_CPUS_PER_TASK:-1}'" echo 'SLURM_NTASKS' $SLURM_NTASKS$ module load Compiler/Intel/18u4 module load Q-Ch/LAMMPS/20Nov2019/intel/2018u4 mpirun -np ${SLURM_NTASKS} lammps -sf opt -pk omp ${OMP_NUM_THREADS} -in $SLURM_JOB_NAME.lam -log $SLURM_JOB_NAME.out >$SLURM_JOB_NAME.stdout === Quantum Espresso === #!/bin/bash #SBATCH -J JOBNAME # job name #SBATCH -N 1 # number of nodes #SBATCH -n 16 # number of cores #SBATCH -o %x.e%j.out # std output file #SBATCH -e %x.e%j.err # std error file #SBATCH --mem-per-cpu=8gb # memory limit module load Compiler/Intel/19u5 module load Q-Ch/QE/6.4.1/intel/2019u5 exe = pw.x # any of qe executables mpirun $exe < scf.in > scf.out