MOLECULAR DYNAMICS PERFORMANCE GUIDE - COMPUTE CANADA

SOFTWARE DETAILS

ID=27, OPENMM.cuda

  • Module/Version: openmm/7.7.0
  • Toolchain/Version: gofbc/2020.1.403.114
  • CPU instruction set: avx2
  • Example job submission script:
    1. #!/bin/bash
    2. #SBATCH -c4 --gpus-per-node=4 -A def-svassili
    3. #SBATCH --mem-per-cpu=4000 --time=1:0:0
    4. # Usage: sbatch submit.cuda.sh
    5. source env-parmed/bin/activate
    6. ml StdEnv/2020 gcc/9.3.0 cuda/11.4 openmpi/4.0.3
    7. ml python/3.8.10 openmm/7.7.0 netcdf/4.7.4 hdf5/1.10.6 mpi4py/3.0.3
    8. python openmm_input.py
  • Benchmark submission script:
    1. #!/bin/bash
    2. #SBATCH -c4 --gpus-per-node=4 -A def-svassili
    3. #SBATCH --mem-per-cpu=4000 --time=1:0:0
    4. # Usage: sbatch submit.cuda.sh
    5. source env-parmed/bin/activate
    6. ml StdEnv/2020 gcc/9.3.0 cuda/11.4 openmpi/4.0.3
    7. ml python/3.8.10 openmm/7.7.0 netcdf/4.7.4 hdf5/1.10.6 mpi4py/3.0.3
    8. echo ${SLURM_NODELIST} running on ${SLURM_CPUS_PER_TASK} cores
    9. cat /proc/cpuinfo | grep "model name" | uniq
    10. python openmm_input.py