batch.sh

#!/bin/bash

## salloc -A m2845 -C gpu -N 1 --gres=gpu:1 -t 2:00:00
#SBATCH -A m2845
#SBATCH -C gpu
#SBATCH -N 1
#SBATCH -t 2:00:00
#SBATCH --gres=gpu:1

echo "==============================="
echo "Running OpenMP offloading benchmark"
echo "==============================="
cat Makefile
srun -n1 ./Benchmark_su3.x --gpu-threads 8
srun -n1 ./Benchmark_su3.x --gpu-threads 32
srun -n1 ./Benchmark_su3.x --gpu-threads 128
#srun -n1 nvprof --print-gpu-trace ./Benchmark_su3.x

#echo "==============================="
#echo "Running reference CUDA benchmark"
#echo "==============================="
#cat Makefile.nvcc

#srun -n1 nvprof ./Benchmark_su3.nvcc.x -o nvcc-8threads.nvp --gpu-threads 8
#srun -n1 nvprof ./Benchmark_su3.nvcc.x -o nvcc-32threads.nvp --gpu-threads 32
#srun -n1 nvprof ./Benchmark_su3.nvcc.x -o nvcc-128threads.nvp --gpu-threads 128

#srun -n1 ./Benchmark_su3.nvcc.x --gpu-threads 64
#srun -n1 ./Benchmark_su3.nvcc.x --gpu-threads 128
#srun -n1 nvprof --print-gpu-trace ./Benchmark_su3.nvcc.x
#srun -n 1 nv-nsight-cu-cli -o omp-cuda-mode ./Benchmark_su3.x