MOLECULAR DYNAMICS PERFORMANCE GUIDE - Digital Research Alliance of CANADA

SOFTWARE DETAILS

ID=28, NAMD2.cuda.ucx

  • Module/Version: namd-ucx-smp/2.14
  • Toolchain/Version: iccifortcuda/2020.1.114
  • CPU instruction set: avx2
  • Example job submission script:
    #!/bin/bash
    #SBATCH --ntasks=1 --nodes=1 -A def-svassili
    #SBATCH --cpus-per-task=12
    #SBATCH --gpus-per-node=1
    #SBATCH --mem-per-cpu=2000 --time=3:0:0
    


    echo numsteps $1 >> run_${SLURM_JOBID}.in
    ml StdEnv/2020 cuda/11.4 namd-ucx-smp/2.14
    


    (( NUM_PES=$SLURM_CPUS_PER_TASK - 1 ))
    srun --mpi=pmi2 namd2 ++ppn $NUM_PES +idlepoll ${RUN_IN} > ${TMPFILE}
  • Benchmark submission script:
      
    #!/bin/bash
    #SBATCH --ntasks=1 --nodes=1 -A def-svassili
    #SBATCH --cpus-per-task=12
    #SBATCH --gpus-per-node=1
    #SBATCH --mem-per-cpu=2000 --time=3:0:0
    # Usage: sbatch $0 number_of_steps
    INPFILE=namd.in
    


    STEPS=$1
    TMPFILE=tf_${SLURM_JOBID}
    RUN_IN=run_${SLURM_JOBID}.in
    cp ${INPFILE} ${RUN_IN}
    echo numsteps $1 >> run_${SLURM_JOBID}.in
    ml StdEnv/2020 cuda/11.4 namd-ucx-smp/2.14
    a
    echo ${SLURM_NODELIST} running on ${SLURM_CPUS_PER_TASK} cores
    cat /proc/cpuinfo | grep "model name" | uniq
    


    (( NUM_PES=$SLURM_CPUS_PER_TASK - 1 ))
    srun --mpi=pmi2 namd2 ++ppn $NUM_PES +idlepoll ${RUN_IN} > ${TMPFILE}