Skip to content

Run

Jump to bottom
Sam Reeve edited this page Aug 5, 2021 · 13 revisions

ORNL CADES

  1. Get an interactive job from a queue with P100 GPUs
salloc -A ccsd -p gpu_p100 --gpus-per-task=1 -N 2 -n4 -c4 -G4 --mem=0G -t 01:00:00 /bin/bash
  1. Load installed python modules and programming environment
source subdir/name-env/bin/activate
module load PE-gnu/3.0
  1. Run example
srun -n 4 python run_config_input.py

ORNL SUMMIT

  1. Get an interactive job (V100 GPUs)
bsub -Is -W 30 -P PROJECT -q batch -nnodes 1 /bin/bash
  1. Load installed python modules and programming environment
module use -a /gpfs/alpine/world-shared/csc143/jyc/summit/sw/modulefiles
module use -a /gpfs/alpine/csc143/world-shared/jyc/summit/sw/spack/share/spack/modules/linux-rhel7-power9le

ml gcc
ml anaconda3
ml py-torch-geometric
  1. Run example
# 1 node, 1 GPU
jsrun -n1 -a1 -c1 -g1 --smpiargs="off" python run_config_input.py
# 4 node, 24 GPU
jsrun -n24 -a1 -g1 -c4 -r6 -b rs --smpiargs="-off" python run_config_input.py
Clone this wiki locally