Follow the following steps to Run Cuda Fortran Example:

Step 1: Download the Makefile local directory and remake it to Makefile.
			Download Fortran file and CUDA file in your local directory.

Step 2: Compile the codes with:

			module load cuda/10.1
			make

Step 3:	Now to run a batch job:

			Create a job submission script "test.bsub", that has:
			#!/bin/bash
			#BSUB -n 1
			#BSUB -q gpu
			#BSUB -J SMG
			#BSUB -W 2
			#BSUB -R "select[rtx2080]"
			#BSUB -gpu "num=1:mode=shared:mps=yes"
			#BSUB -o out.%J
                        #BSUB -e err.%J
                        module load PrgEnv-intel/2018.2.199
			module load cuda/10.1
			mpirun -n 1 ./a.out

			Submit the job with "bsub < test.bsub".
			
			This job will run on the gpu queue (#BSUB -q gpu).
			Also, it will run on the newest NVIDIA RTX2080 GPU node, which is specified with 
			#BSUB -R "select[rtx2080]" -gpu "num=1:mode=shared:mps=yes"
			To see all the node labels in the gpu queue, use "lshosts | grep gpu".