adding defiant2 mpi example

Author: secondspass

defiant/mpiexample/README.md

@@ -0,0 +1,16 @@
+# MPI Example
+
+This example shows how to install MPICH 4.1.2 in a container derived from a CUDA 12.6 container released by Nvidia. 
+
+Build the images by running 
+
+```
+./build.sh
+```
+
+Submit the batch script
+
+```
+sbatch submit.sl
+```
+

defiant/mpiexample/build.sh

@@ -0,0 +1,4 @@
+#!/bin/bash
+
+apptainer build rocky9opensusempich412nvidia2411.sif rocky9opensusempich412nvidia2411.def
+apptainer build mpiexample.sif mpiexample.def

defiant/mpiexample/mpiexample.c

@@ -0,0 +1,19 @@
+#include <stdio.h>
+#include <mpi.h>
+
+int main (int argc, char *argv[])
+{
+int rank, size;
+MPI_Comm comm;
+
+comm = MPI_COMM_WORLD;
+MPI_Init (&argc, &argv);
+MPI_Comm_rank (comm, &rank);
+MPI_Comm_size (comm, &size);
+
+printf("Hello from rank %d\n", rank);
+
+MPI_Barrier(comm);
+MPI_Finalize();
+}
+

defiant/mpiexample/mpiexample.def

@@ -0,0 +1,10 @@
+Bootstrap: localimage
+From: rocky9opensusempich412nvidia2411.sif
+
+%files
+mpiexample.c /app/mpiexample.c
+
+
+%post
+cd /app && mpicc -o mpiexample mpiexample.c
+

defiant/mpiexample/rocky9opensusempich412nvidia2411.def

@@ -0,0 +1,34 @@
+Bootstrap: docker
+From: nvcr.io/nvidia/nvhpc:24.11-devel-cuda12.6-rockylinux9
+
+%environment
+    # Point to MPICH binaries, libraries man pages
+    export MPICH_DIR=/opt/mpich
+    export PATH="$MPICH_DIR/bin:$PATH"
+    export LD_LIBRARY_PATH="$MPICH_DIR/lib:$LD_LIBRARY_PATH"
+    export MANPATH=$MPICH_DIR/share/man:$MANPATH
+
+%post
+
+
+# Information about the version of MPICH to use
+export MPICH_VERSION=4.1.2
+export MPICH_URL="http://www.mpich.org/static/downloads/$MPICH_VERSION/mpich-$MPICH_VERSION.tar.gz"
+export MPICH_DIR=/opt/mpich
+
+echo "Installing MPICH..."
+rm -rf /tmp/mpich
+mkdir -p /tmp/mpich
+mkdir -p /opt
+# Download
+cd /tmp/mpich && wget -O mpich-$MPICH_VERSION.tar.gz $MPICH_URL && tar --no-same-owner -xzf mpich-$MPICH_VERSION.tar.gz
+# Compile and install
+export FFLAGS=-fallow-argument-mismatch
+
+cd /tmp/mpich/mpich-$MPICH_VERSION 
+#patch -p1 < /my_change.patch
+#./autogen.sh
+./configure  --with-device=ch4:ucx --prefix=$MPICH_DIR && make install
+rm -rf /tmp/mpich
+
+

defiant/mpiexample/submit.sl

@@ -0,0 +1,21 @@
+#!/bin/bash 
+#SBATCH -t00:10:00 
+#SBATCH -A stf007
+#SBATCH -p batch-cpu
+#SBATCH -N2
+#SBATCH -J mpiexample
+#SBATCH -o logs/%x_%j.out
+#SBATCH -e logs/%x_%j.out
+
+module reset
+module load cray-mpich-ucx-abi
+
+export MPICH_SMP_SINGLE_COPY_MODE=NONE
+export APPTAINERENV_LD_LIBRARY_PATH="/opt/cray/pe/mpich/8.1.32/ucx/crayclang/18.0/lib-abi-mpich:/opt/cray/pe/mpich/8.1.32/gtl/lib:$CRAY_LD_LIBRARY_PATH:$LD_LIBRARY_PATH:/opt/cray/pe/lib64:/usr/lib64/libibverbs:/opt/cray/pals/1.6/lib:/usr/lib64/ucx"
+export APPTAINER_CONTAINLIBS="/usr/lib64/libjansson.so.4,/usr/lib64/libdrm.so.2,/lib64/libtinfo.so.6,/usr/lib64/libnl-3.so.200,/usr/lib64/librdmacm.so.1,/usr/lib64/libibverbs.so.1,/usr/lib64/libibverbs/libmlx5-rdmav57.so,/lib64/libucs.so.0,/lib64/libucp.so.0,/lib64/libuct.so.0,/lib64/libucm.so.0"
+export APPTAINER_BIND=/usr/share/libdrm,/var/spool/slurm,/opt/cray,/opt/mellanox,/etc/libibverbs.d,/usr/lib64/ucx,${PWD}
+
+
+export CONTAINER=../containers/mpiexample.sif
+
+srun --unbuffered -N2 -n4 --tasks-per-node 2 apptainer exec --fakeroot  --workdir `pwd` $CONTAINER /app/mpiexample