Installation on Perlmutter (NERSC)#

Installation#

Execute the following commands in order to create a dedicated Python environment, in which to install optimas.

module load cray-python/3.9.13.1

python3 -m pip install --user --upgrade pip
python3 -m pip install --user virtualenv
python3 -m pip cache purge
python3 -m venv $HOME/sw/perlmutter/gpu/venvs/optimas
source $HOME/sw/perlmutter/gpu/venvs/optimas/bin/activate

pip install 'optimas[all]'

Running an optimas job#

In order to run a new optimas job, create a new folder in the $SCRATCH directory, and add the files needed to describe your optimization, i.e. typically:

a file run_optimization.py
a template simulation input script
a file analysis_script.py that postprocesses the simulation results

(See a typical example here.)

Then, create a file submission_script with the following content:

#!/bin/bash -l
#SBATCH -t <walltime>
#SBATCH -N <n_nodes>
#SBATCH -A <nersc_account>
#SBATCH -C gpu
#SBATCH -q regular
#SBATCH --exclusive
#SBATCH --gpu-bind=none
#SBATCH --gpus-per-node=4

module load cray-python/3.9.13.1
source $HOME/sw/perlmutter/gpu/venvs/optimas/bin/activate

python run_optimization.py

where <walltime>, <n_nodes> and <nersc_account> should be replaced by the wall time, number of nodes and NERSC account number that you want to use.

Then run:

sbatch submission_script