WRF

General

The Weather Research and Forecasting (WRF) Model is a next-generation mesoscale numerical weather prediction system designed to serve both operational forecasting and atmospheric research needs.

Description

WRF features multiple dynamical cores, a 3-dimensional variational (3DVAR) data assimilation system, and a software architecture allowing for computational parallelism and system extensibility. WRF is suitable for a broad spectrum of applications across scales ranging from meters to thousands of kilometers.

Usage at HPC2N

At HPC2N we have currently installed WRF and WPS (WRF Preprocessing System)

The binaries of WRF/WPS are available through the module system.
To access do module add wrf on the command line and/or in your submit file.

Since the actual wrf binary can be compiled in several different ways we have done so and the binaries are therefor named according to how they where built.

• wrf.basic.sm+dm.mpi - built with both OpenMP and MPI with basic nesting
• wrf.move_nests.sm+dm.mpi - built with both OpenMP and MPI with nesting set to preset moves

If these aren't sufficient please inform support@hpc2n.umu.se of what you need and we will build it.

All other binaries from WRF are available in normal serial versions.

The input tables are located under /lap/wrf/data/...

Submitfile examples

Since wrf is built as a OpenMP/MPI combined binary special care must be taken in the submitfile.

#/bin/bash
# Only request full nodes
#PBS -l nodes=8
# To get full nodes request all available memory per node
#PBS -l pmem=15000m
#PBS -l pvmem=25000m
#PBS -l walltime=6:00:00

cd $PBS_O_WORKDIR

# Increase stack limit since WRF consumes lots of stack space
ulimit -s 1048576

module add wrf
# Use all 8 cores on each node
export OPENMP_NUM_THREADS=8

mpiexec wrf.basic.sm+dm.mpi

Additionial info

Documentation is available on the WRF homepage and the user guide