Page Comparison

...

1. Copy build_examples/build_cpu_mpi-only_intel_ubuntu20.04.sh to root directory.

   • cp build_examples/build_cpu_mpi-only_intel_ubuntu20.04.sh rebuild.sh

2. Build HDF5 and edit rebuild.sh with HDF5 location.

3. FC = mpif90” FC = $(MPIFC)

4. Execute the build script.

Sample build script for Fritz:

We used modules for MPI and HDF5 on Firtz.

module load intel/2022# MPI=openmpi-4.1.2
module load compiler/2022.1.0
module load mkl/2022.0.2
# module load impi/2021.6.0
module load hpcx/2.13.0

-intel2021.4.0
MPI=intelmpi-2021.7.0
if [[ "$MPI" =~ impi^intel ]]; then
        exportmodule MPIFC=mpiifort
elif [[ "$MPI" =~ hpcx ]]; thenload hdf5/1.10.7-impi-intel
          export OMPII_MPI_MPICCF90=iccifort
else
       export OMPI_MPICXX=icpcmodule         export OMPI_MPIFC=ifortload hdf5/1.10.7-ompi-intel
           export OMPI_MPIF90=ifort
      
 export MPIFC=mpif90
fi

module#################################################################
load
hdf5/1.12.1-$MPI

# Location of local hdf5 installed with same compiler being used for POT3D:
HDF5_INCLUDE_DIR="$HDF5_DIRROOT/include"
HDF5_LIB_DIR="$HDF5_DIRROOT/lib"
# Fortran HDF5 library flags (these can be version dependent):
HDF5_LIB_FLAGS="-lhdf5_fortran -lhdf5hllhdf5_hl_fortran -lhdf5 -lhdf5_hl""
...

Running Example

cd testsuite
TEST=small
POT3D_HOME=$PWD/..

cp ${POT3D_HOME}/testsuite/${TEST}/input/* ${POT3D_HOME}/testsuite/${TEST}/run/
cd ${POT3D_HOME}/testsuite/${TEST}/run

echo "Running POT3D with $NP MPI rank..."
CMD="mpirun -np $NP $MPIFLAGS $POT3D_HOME/bin/pot3d"

/usr/bin/time $CMD > pot3d.out
echo "Done!"

runtime=($(tail -n 5 timing.out | head -n 1))
echo "Wall clock time: ${runtime[6]} seconds"
echo " "

#Validate run:
${POT3D_HOME}/scripts/pot3d_validation.sh pot3d.out ${POT3D_HOME}/testsuite/${TEST}/validation/pot3d.out

...

1. Profile the input
   Use any of the remote clusters to run an MPI profile (such as IPM profile or any other profiler) over 4 nodes, full PPN to profile the given input.

   • Submit the profile as PDF to the team's folder.

   • Add to your presentation the 3 main MPI calls that are being used plus the MPI time being used.

2. Run POT3D with isc2023 input on both PSC bridges-2 and FAU Fritz CPU clusters.

   Code Block
   Your job should converge and print statementsoutputs like below. ### The CG solver has converged. Iteration: 25112 Residual: 9.972489313728662E-13

3. Run POT3D on the FAU Alex cluster using A100 GPU partition. Use only 4 GPUs for the run. Submit the results to the team's folder.

4. Submission and Presentation:
   - Submit all the build scripts, run scripts and input & outputs (pot3d.dat, pot3d.out, timing.out)
   - Do not submit the output data or the source code.
   - Prepare slides for the team’s interview based on your work for this application.

...