In this project we'll try lots of frameworks for using Python on GPUs.
Get shared node on cori gpu:
module load esslurm python cuda
salloc -C gpu -N 1 -t 60 -c 10 -G 1 -A m1759 -q special
./bash_orchestrator.sh
Modify bash_orchestrator.sh to choose frameworks and array sizes.
Results, timing data, and logfile are written to /results directory.
The logfile is called log_${SLURMJOBID}.out
The bash_orchestrator.sh script coordinates the benchmark. This was required
in order to be able to hop in and out of conda environments for each framework.
The run_benchmark.py takes care of the rest inside each conda
environment/framework.
In each framework directory the framework_requirements.txt file details the
contents of each conda environment.