Submitting job scripts

1. Environment
2. Example of a QP job script
3. Example of a CHAMP job script
- 3.1. Variational Monte Carlo run
- 3.2. Diffusion Monte Carlo run

1 Environment

Some environment variables should be set for the codes to run properly. To set up the environment, you should source the ~scemama/environment.sh file in your shell when you log in:

source ~scemama/environment.sh

This file contains the following commands:

module load openblas                        # QP was compiled with OpenBlas
source ~scemama/qp2/quantum_package.rc      # QP environment
source ~filippi/intel/oneapi/setvars.sh     # Intel libraries for CHAMP

It is important also that you install the following Python packages

pip install trexio
pip install resultsFile

2 Example of a QP job script

QP uses one process, and all possible threads on the node. You should request an exclusive node with one process:

#!/bin/bash
#SBATCH -N 1 -n 1  --exclusive
#SBATCH --account=reserved
#SBATCH --partition=reserved

source ~scemama/environment.sh

qp set_file COH2
qp run fci > COH2.fci.out

When you work with SLURM, you should use sbatch to submit your scripts: #+begin_src bash sbatch job.sh #+end_src bash

3 Example of a CHAMP job script

CHAMP is parallelized with MPI, and doesn't use multi-threading. You should request a single node, with 16 processes (as there are 16 cores per node).

3.1 Variational Monte Carlo run

#!/bin/bash
#SBATCH -N 1 -n 16
#SBATCH --account=reserved
#SBATCH --partition=reserved

source ~scemama/environment.sh

INPUT=vmc_h2o_hf.inp

OUTPUT=${INPUT%.inp}.out
mpirun -np 16 ~filippi/champ-2.1.0/bin/vmc.mov1  -i $INPUT -o $OUTPUT -e error

3.2 Diffusion Monte Carlo run

#!/bin/bash
#SBATCH -N 1 -n 16

source ~scemama/environment.sh

INPUT=dmc_h2o_hf.inp

OUTPUT=${INPUT%.inp}.out
mpirun -np 16 ~filippi/champ-2.1.0/bin/dmc.mov1  -i $INPUT -o $OUTPUT -e error