Polaris
========
If this is your first time building MFIX-Exa on Polaris, please
review the general notes below and `Basics`_ section first.
* The documentation for this system can be found `here. `_
* Polaris can be accessed from a system with ssh client installed. These connections also work from NETL's SciLAN and Joule:
.. code:: bash
ssh @polaris.alcf.anl.gov
Login with your Mobilepass token. No need to prefix or append with anything.
* These instructions build MFIX-Exa on the login nodes using `-j8` CPUs.
You may have to decrease this value if there is high traffic
or you may want to increase this value if you are on a compute
node interactively.
* The cmake instructions compile to a `build` directory.
The gmake instructions compile to a `exec` directory.
* For the dependencies, it is assumed that you have set the
following environment variables:
.. code:: bash
export HYPRE_INSTALL_DIR=$HOME/
export CSG_INSTALL_DIR=$HOME/
export CSG_LIB_DIR=$HOME/
export ASCENT_INSTALL_DIR=$HOME/
to a path that you have read/write access to,
such as inside the ``/lus/grand/projects//`` space corresponding to your account.
You will need to recall these paths later if you want to build
MFIX-Exa with the optional dependencies.
* After building the `mfix` executable (with cmake), you can
build the PIC-to-DEM restarter app by executing the following command
in the `build` directory
.. code:: bash
cmake --build . --target pic2dem
Basics
------
Source code
~~~~~~~~~~~
Before building, first obtain a copy of the source code
following the instructions on the
`MFIX-Exa website. `_
Modules
~~~~~~~
All of the build instructions below have been tested with the
following modules and environment helpers
.. code:: bash
module swap PrgEnv-nvhpc PrgEnv-gnu
module load cmake/3.23.2
The GPU-enabled builds additionally require
.. code:: bash
module load nvhpc-mixed
export MPICH_GPU_SUPPORT_ENABLED=1
export AMREX_CUDA_ARCH=8.0
Full builds that utilize external dependencies, also require setting
certain environment variables as discussed below.
Building MFIX-Exa
-----------------
The commands below are the superbuild instructions, i.e.,
AMReX is built as part of the MFIX-Exa build process.
To build MFIX-Exa with hypre, csg and/or ascent dependencies,
you first need to build and install these libraries and their dependencies.
Instructions on building the necessary dependencies are below
and should be successfully installed first. There are two primary
methods of building MFIX-Exa: `cmake` and `gmake` which are provided
separately below.
cmake
~~~~~
.. tabs::
.. tab:: CPU
.. code:: bash
cmake -DCMAKE_C_COMPILER=gcc \
-DCMAKE_CXX_COMPILER=g++ \
-DCMAKE_Fortran_COMPILER=gfortran \
-DMFIX_MPI=yes \
-DMFIX_OMP=no \
-DMFIX_GPU_BACKEND=NONE \
-DAMReX_TINY_PROFILE=no \
-DMFIX_CSG=no \
-DMFIX_HYPRE=no \
-DCMAKE_BUILD_TYPE=Release \
../
make -j8
.. tab:: GPU
.. code:: bash
cmake -DCMAKE_C_COMPILER=gcc \
-DCMAKE_CXX_COMPILER=g++ \
-DCMAKE_Fortran_COMPILER=gfortran \
-DMFIX_MPI=yes \
-DMFIX_OMP=no \
-DMFIX_CSG=no \
-DMFIX_HYPRE=no \
-DMFIX_GPU_BACKEND=CUDA \
-DGPUS_PER_NODE=4 \
-DAMReX_TINY_PROFILE=no \
-DCMAKE_BUILD_TYPE=Release \
../
make -j8
.. tab:: CPU-full
.. code:: bash
export HYPRE_DIR=$HYPRE_INSTALL_DIR
export HYPRE_ROOT=$HYPRE_DIR
export HYPRE_LIBRARIES=$HYPRE_DIR/lib
export HYPRE_INCLUDE_DIRS=$HYPRE_DIR/include
export ASCENT_DIR=$ASCENT_INSTALL_DIR
export CONDUIT_DIR=$ASCENT_DIR
export CMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH:$ASCENT_DIR/lib/cmake/ascent
export CMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH:$ASCENT_DIR/lib/cmake/conduit
export CMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH:$CSG_INSTALL_DIR
cmake -DCMAKE_C_COMPILER=gcc \
-DCMAKE_CXX_COMPILER=g++ \
-DCMAKE_Fortran_COMPILER=gfortran \
-DMFIX_MPI=yes \
-DMFIX_OMP=no \
-DMFIX_CSG=yes \
-DMFIX_HYPRE=yes \
-DAMReX_ASCENT=yes \
-DAMReX_CONDUIT=yes \
-DMFIX_GPU_BACKEND=NONE \
-DAMReX_TINY_PROFILE=no \
-DCMAKE_BUILD_TYPE=Release \
../
make -j8
.. tab:: GPU-full
.. code:: bash
export HYPRE_DIR=$HYPRE_INSTALL_DIR
export HYPRE_ROOT=$HYPRE_DIR
export HYPRE_LIBRARIES=$HYPRE_DIR/lib
export HYPRE_INCLUDE_DIRS=$HYPRE_DIR/include
export ASCENT_DIR=$ASCENT_INSTALL_DIR
export CONDUIT_DIR=$ASCENT_DIR
export CMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH:$ASCENT_DIR/lib/cmake/ascent
export CMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH:$ASCENT_DIR/lib/cmake/conduit
export CMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH:$CSG_INSTALL_DIR
cmake -DCMAKE_C_COMPILER=gcc \
-DCMAKE_CXX_COMPILER=g++ \
-DCMAKE_Fortran_COMPILER=gfortran \
-DMFIX_MPI=yes \
-DMFIX_OMP=no \
-DMFIX_CSG=yes \
-DMFIX_HYPRE=yes \
-DAMReX_ASCENT=yes \
-DAMReX_CONDUIT=yes \
-DMFIX_GPU_BACKEND=CUDA \
-DGPUS_PER_NODE=4 \
-DAMReX_TINY_PROFILE=no \
-DCMAKE_BUILD_TYPE=Release \
../
make -j8
Optional build dependencies
---------------------------
The following dependencies need to be built and installed
prior to following any of the full build instructions above.
#. Set environment helpers
.. code:: bash
export CC=$(which cc)
export CXX=$(which CC)
mkdir $HOME/scratch && cd $HOME/scratch
#. HYPRE
The GPU build for HYPRE seems to only work with ``cmake``.
.. tabs::
.. tab:: CPU
.. code:: bash
git clone https://github.com/hypre-space/hypre.git
pushd hypre/src/
git checkout v2.26.0
./configure --prefix=$HYPRE_INSTALL_DIR --with-MPI
make -j8 install
popd
.. tab:: GPU
.. code:: bash
git clone https://github.com/hypre-space/hypre.git
pushd hypre/src/
git checkout v2.26.0
cmake -S . -B build \
-DCMAKE_INSTALL_PREFIX=$HYPRE_INSTALL_DIR \
-DHYPRE_WITH_DSUPERLU=OFF \
-DHYPRE_ENABLE_BIGINT=OFF \
-DHYPRE_WITH_OPENMP=OFF \
-DHYPRE_ENABLE_SHARED=ON \
-DHYPRE_WITH_MPI=ON \
-DHYPRE_WITH_CUDA=ON \
-DCMAKE_CUDA_ARCHITECTURES=80 \
-DHYPRE_ENABLE_CUSPARSE=ON \
-DHYPRE_ENABLE_CURAND=ON \
-DCMAKE_CXX_COMPILER=$(which nvc++) \
-DMPI_CXX_COMPILER=$(which CC) \
-DCMAKE_C_COMPILER=$(which nvc) \
-DMPI_C_COMPILER=$(which cc) \
-DHYPRE_INSTALL_PREFIX=$HYPRE_INSTALL_DIR
cd build/
make -j8 install
popd
#. Catch2
.. code:: bash
git clone --depth 1 --branch v3.11.0 https://github.com/catchorg/Catch2
pushd Catch2/
cmake -S . -B build -DCMAKE_INSTALL_PREFIX=$CSG_INSTALL_DIR
cd build/
make -j8 install
popd
#. GMP
.. code:: bash
wget --no-check-certificate https://ftp.gnu.org/gnu/gmp/gmp-6.2.1.tar.xz
tar -xf gmp-6.2.1.tar.xz
pushd gmp-6.2.1
./configure --prefix=$CSG_INSTALL_DIR
make -j8 install
popd
#. MPFR
.. code:: bash
wget --no-check-certificate https://ftp.gnu.org/gnu/mpfr/mpfr-4.1.0.tar.xz
tar -xf mpfr-4.1.0.tar.xz
pushd mpfr-4.1.0/
./configure --with-gmp=$CSG_INSTALL_DIR --prefix=$CSG_INSTALL_DIR
make -j8 install
popd
#. Boost
.. code:: bash
wget https://boostorg.jfrog.io/artifactory/main/release/1.81.0/source/boost_1_81_0.tar.gz
tar -zxvf boost_1_81_0.tar.gz
pushd boost_1_81_0/
./bootstrap.sh
./b2 install --prefix=$CSG_INSTALL_DIR
popd
#. CGAL
.. code:: bash
git clone --depth 1 --branch v6.1.1 https://github.com/CGAL/cgal
pushd cgal/
cmake -S . -B build -DCMAKE_INSTALL_PREFIX=$CSG_INSTALL_DIR
cd build/
make -j8 install
popd
#. PEGTL
.. code:: bash
git clone --branch 3.2.8 https://github.com/taocpp/PEGTL
pushd PEGTL/
cmake -S . -B build -DCMAKE_INSTALL_PREFIX=$CSG_INSTALL_DIR
cd build/
make -j8 install
popd
#. Conduit
.. code:: bash
git clone --recursive https://github.com/LLNL/conduit.git
pushd conduit/
git checkout v0.8.6
mkdir build && cd build
cmake -S ../src -DCMAKE_INSTALL_PREFIX=$ASCENT_INSTALL_DIR \
-DENABLE_OPENMP=OFF \
-DENABLE_MPI=ON \
-DENABLE_CUDA=OFF \
-DCMAKE_BUILD_TYPE=Release
make -j8 install
popd
#. Vtk-m
.. code:: bash
git clone --branch master https://gitlab.kitware.com/vtk/vtk-m.git
pushd vtk-m/
git checkout v1.9.0
mkdir build && cd build/
cmake -S ../ -DCMAKE_INSTALL_PREFIX=$ASCENT_INSTALL_DIR \
-DVTKm_ENABLE_OPENMP=OFF \
-DVTKm_ENABLE_MPI=ON \
-DVTKm_ENABLE_CUDA=OFF \
-DVTKm_USE_64BIT_IDS=OFF \
-DVTKm_USE_DOUBLE_PRECISION=ON \
-DVTKm_USE_DEFAULT_TYPES_FOR_ASCENT=ON \
-DVTKm_NO_DEPRECATED_VIRTUAL=ON \
-DCMAKE_BUILD_TYPE=Release
make -j8 install
popd
#. Ascent
.. code:: bash
git clone --recursive https://github.com/Alpine-DAV/ascent.git
pushd ascent
git checkout v0.9.0
mkdir build && cd build/
cmake -S ../src -DCMAKE_INSTALL_PREFIX=$ASCENT_INSTALL_DIR \
-DCONDUIT_DIR=$ASCENT_INSTALL_DIR \
-DVTKM_DIR=$ASCENT_INSTALL_DIR \
-DENABLE_VTKH=ON \
-DENABLE_FORTRAN=OFF \
-DENABLE_PYTHON=OFF \
-DENABLE_DOCS=OFF \
-DBUILD_SHARED_LIBS=ON \
-DCMAKE_BUILD_TYPE=Release \
-DENABLE_GTEST=OFF \
-DENABLE_TESTS=OFF
make -j8 install
popd
Running Jobs
------------
Common Slurm commands:
* **qsub runit.sh** submit a job to the queue
* **qstat -u USER** check job status of user USER
* **qdel JOBID** kill a job with id JOBID
* **qsub -I -l select=1:system=polaris -l walltime=0:60:00 -l filesystems=home:gran -q debug -A ** grab 1 GPU node (for up to 1 hrs)
Example run scripts:
Example run script for GPU is below. For CPU-only,
remove ``module load nvhpc-mixed`` and ``export MPICH_GPU_SUPPORT_ENABLED=1``,
and the options ``ppn`` and ``--cpu-bind core`` from the run line.
.. code:: bash
#!/bin/bash
#PBS -A
#PBS -q prod
#PBS -l select=64:system=polaris
#PBS -l filesystems=home:eagle
#PBS -l walltime=04:00:00
#PBS -N alcc_tk1
#PBS -k doe
#PBS -o stdout
#PBS -e stderr
#PBS -m be
##PBS -M # Be default, mail goes to the submitter, use this option to add others (uncomment to use)
NNODES=`wc -l < $PBS_NODEFILE` # number of nodes requested
RPN=4 # assume 1 GPU = 1 resource, 4 GPUs / node
RES=$((NNODES*RPN)) # total number of resources
echo 'procs = '$procs' '
echo '$PBS_O_WORKDIR = '$PBS_O_WORKDIR' '
cd $PBS_O_WORKDIR
module swap PrgEnv-nvhpc PrgEnv-gnu
module load nvhpc-mixed
export MPICH_GPU_SUPPORT_ENABLED=1
set +x # report all commands to stderr
env # save the env
mpiexec -n $RES -ppn $RPN --cpu-bind core ./mfix inputs > screen.1
wait 3
mpiexec -n $RES -ppn $RPN --cpu-bind core ./mfix inputs > screen.2
wait 7
mpiexec -n $RES -ppn $RPN --cpu-bind core ./mfix inputs > screen.3
echo 'eof EOF EOR eor'