Format of an execution script
===================================

This section describes the format of execution script files to run programs on the supercomputer. To execute the application that requires an execution script file, create the file in advance.  


Execute a non MPI program
----------------------------------------

.. code-block :: none

   #!/bin/sh 
   #PBS -l select=1 
   #PBS -q queue name
   #PBS -N job name  

   cd $PBS_O_WORKDIR
   
   program  > output file 2> error file 

・Example: To execute a program ‘a.out’.

.. code-block :: none

   #!/bin/sh 
   #PBS -l select=1 
   #PBS -q CP_001 
   #PBS -N sample 

   cd $PBS_O_WORKDIR
   
   ./a.out > result.out 2> result.err 
   

Execute a MPI program 
------------------------------------

.. code-block :: none

   #!/bin/sh 
   #PBS -l select=nodes:ncpus=cores:mem=XXgb
   #PBS -l walltime=HH:MM:SS
   #PBS -q queue
   #PBS -N jobname

   cd $PBS_O_WORKDIR

   mpirun [ -np MPI total tasks | -ppn MPI tasks per node ] program > output file 2> error file 

・Example: To run a program on shared node using Intel MPI, with 8 MPI processes, 32GB memory, and walltime of 1 hour.

.. code-block :: none

   #!/bin/sh 
   #PBS -l select=1:ncpus=8:mem=32gb
   #PBS -l walltime=01:00:00
   #PBS -q CP_001
   #PBS -N mpi 

   module load oneapi
   cd $PBS_O_WORKDIR

   mpirun -np 8 ./a.out > result.out 2> result.err 

・Example: To run a program on 2 node using Intel MPI, with 112 MPI processes each nodes.

.. code-block :: none

   #!/bin/sh 
   #PBS -l select=2 
   #PBS -q P_030
   #PBS -N mpi 

   module load oneapi
   cd $PBS_O_WORKDIR

   mpirun -np 224 -ppn 112 ./a.out > result.out 2> result.err 


Execute a GPU program
------------------------------------

.. code-block :: none

   #!/bin/sh 
   #PBS -l select=nodes:ncpus=cores:mem=XXgb
   #PBS -l walltime=HH:MM:SS
   #PBS -q queue
   #PBS -N jobname

   cd $PBS_O_WORKDIR

   mpirun [ -np MPI total tasks | -ppn MPI tasks per node ] program > output file 2> error file 


・Example: To run a program on shared node using NVIDIA HPC SDK, with 2 MPI processes, 2GPUs, 64GB memory, and walltime of 30 minutes.

.. code-block :: none

   #!/bin/sh 
   #PBS -l select=1:ncpus=4:ngpus=4:mem=64gb
   #PBS -l walltime=00:30:00
   #PBS -q CA_001
   #PBS -N mpi 

   module load nvhpc
   cd $PBS_O_WORKDIR

   mpirun -np 4 -hostfile $PBS_NODEFILE -x LD_LIBRARY_PATH -x HCOLL_MAIN_IB=all ./a.out > result.out 2> result.err