Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Excerpt
hiddentrue
nameExample 1: Run WRF

ตัวอย่างการใช้งาน WRF เบื้องต้น

...

Code Block
languagebash
#!/bin/bash
#SBATCH -p compute                  # Partition or machine type [devel/compute/memory]
#SBATCH -N 1 --ntasks-per-node=40        # Number of nodes and Number of core per node #SBATCH -t 24:00:00    # Number of nodes
#SBATCH --ntasks-per-node=40        # TotalNumber runof timeMPI limit (hourprocesses per node
#SBATCH --cpus-per-task=1           # Number of OpenMP threads per MPI process
#SBATCH -t 24:00:00                 # Total run time limit (hour:minute:second)
#SBATCH -J WRF_Ex1                  # Job name (short)
#SBATCH -A projxxxx                 # Your project account *** {USER EDIT} ***

module purge
module load netCDF-Fortran/4.5.2-iimpi-2019b
module load libpng/1.6.37-GCCcore-8.3.0
module load JasPer/1.900.1-intel-2019b

export OMP_NUM_THREADS=1            # OpenMP threads per MPI process${SLURM_CPUS_PER_TASK}
export JASPERINC=$EBROOTJASPER/include
export JASPERLIB=$EBROOTJASPER/lib

ulimit -s unlimited                 # MUST have otherwise ERROR

srun -n8 ./real.exe                 # Run real.exe using 8 CPU cores (-n8)
srun -n40 -c1n${SLURM_NTASKS} -c${SLURM_CPUS_PER_TASK} ./wrf.exe

# It is more efficient to run geogrid.exe, metgrid.exe and real.exe using a few CPU cores (but enough memory).

...

Code Block
languagebash
sbatch submitWRF.sh
tail -F rsl.out.0000        # (Optional) see the progress

RETURN: WRF model

...

Contact Us
ThaiSC support service : thaisc-support@nstda.or.th