MPI on B4F cluster - Revision history

Haars0011: Fixed first part of markup

2023-10-26T11:24:42Z

Fixed first part of markup

← Older revision		Revision as of 13:24, 26 October 2023
Line 3:		Line 3:
	Consider the following simple MPI version, in C, of the 'Hello World' example:		Consider the following simple MPI version, in C, of the 'Hello World' example:

	~~<source lang='cpp'>~~		#include <stdio.h>
	#include <stdio.h>		#include <mpi.h>
	#include <mpi.h>		int main(int argc, char ** argv) {
	int main(int argc, char ** argv) {		int size,rank,namelen;
	int size,rank,namelen;		char processor_name[MPI_MAX_PROCESSOR_NAME];
	char processor_name[MPI_MAX_PROCESSOR_NAME];		MPI_Init(&argc, &argv);
	MPI_Init(&argc, &argv);		MPI_Comm_rank(MPI_COMM_WORLD,&rank);
	MPI_Comm_rank(MPI_COMM_WORLD,&rank);		MPI_Comm_size(MPI_COMM_WORLD,&size);
	MPI_Comm_size(MPI_COMM_WORLD,&size);		MPI_Get_processor_name(processor_name, &namelen);
	MPI_Get_processor_name(processor_name, &namelen);		printf("Hello MPI! Process %d of %d on %s\n", rank, size, processor_name);
	printf("Hello MPI! Process %d of %d on %s\n", rank, size, processor_name);		MPI_Finalize();
	MPI_Finalize();		}
	}
	~~</source>~~

	Before compiling, make sure that the compilers that are required available.		Before compiling, make sure that the compilers that are required available.
	~~<pre>~~		module list
	module list
	~~</pre>~~

	To avoid conflicts between libraries, the safest way is purging all modules:		To avoid conflicts between libraries, the safest way is purging all modules:
	~~<source lang='bash'>~~		module purge
	module purge
	~~</source>~~

	~~The~~ load both gcc and openmpi libraries. If modules were purged, then slurm needs to be reloaded too.		Then load both gcc and openmpi libraries. If modules were purged, then slurm needs to be reloaded too.
	~~<source lang='bash'>~~		module load gcc openmpi/gcc slurm
	module load gcc~~/4.8.1~~ openmpi/gcc~~/64/1.6.5~~ slurm~~/2.5.7~~
	~~</source>~~

	Compile the <code>hello_mpi.c</code> code.		Compile the <code>hello_mpi.c</code> code.
	~~<source lang='bash'>~~		mpicc hello_mpi.c -o test_hello_world
	mpicc hello_mpi.c -o test_hello_world
	~~</source>~~

	If desired, a list of libraries compiled into the executable can be viewed:		If desired, a list of libraries compiled into the executable can be viewed:
	~~<source lang='bash'>~~		ldd test_hello_world
	ldd test_hello_world
	~~</source>~~

	linux-vdso.so.1 => (~~0x00002aaaaaacb000~~)		linux-vdso.so.1 (0x00007ffc6fb18000)
	~~libmpi~~.so.1 => /~~shared~~/~~apps~~/~~openmpi~~/~~gcc~~/64/1.6.5/~~lib64~~/~~libmpi~~.so.1 (~~0x00002aaaaaccd000~~)		libmpi.so.40 => /usr/lib/x86_64-linux-gnu/libmpi.so.40 (0x000014d19dfb2000)
	~~libdl~~.so.2 => /~~lib64~~/~~libdl~~.so.2 (~~0x00002aaaab080000~~)		libpthread.so.0 => /usr/lib/x86_64-linux-gnu/libpthread.so.0 (0x000014d19df8f000)
	libm.so.6 => /~~lib64~~/libm.so.6 (~~0x00002aaaab284000~~)		libc.so.6 => /usr/lib/x86_64-linux-gnu/libc.so.6 (0x000014d19dd9d000)
	~~libnuma~~.so.1 => /usr/lib64/~~libnuma~~.so.1 (~~0x0000003e29400000~~)		libopen-rte.so.40 => /usr/lib/x86_64-linux-gnu/libopen-rte.so.40 (0x000014d19dce3000)
	~~librt~~.so.1 => /~~lib64~~/~~librt~~.so.1 (~~0x00002aaaab509000~~)		libopen-pal.so.40 => /usr/lib/x86_64-linux-gnu/libopen-pal.so.40 (0x000014d19dc33000)
	~~libnsl~~.so.1 => /~~lib64~~/~~libnsl~~.so.1 (~~0x00002aaaab711000~~)		libm.so.6 => /usr/lib/x86_64-linux-gnu/libm.so.6 (0x000014d19dae4000)
	libutil.so.1 => /~~lib64~~/libutil.so.1 (~~0x00002aaaab92a000~~)		libhwloc.so.15 => /usr/lib/x86_64-linux-gnu/libhwloc.so.15 (0x000014d19da93000)
	~~libpthread~~.so.0 => /~~lib64~~/~~libpthread~~.so.0 (~~0x00002aaaabb2e000~~)		/lib64/ld-linux-x86-64.so.2 (0x000014d19e0d9000)
	~~libc~~.so.6 => /~~lib64~~/~~libc~~.so.6 (~~0x00002aaaabd4b000~~)		libz.so.1 => /usr/lib/x86_64-linux-gnu/libz.so.1 (0x000014d19da77000)
	/~~lib64~~/ld-linux-~~x86-64~~.so.2 (~~0x00002aaaaaaab000~~)		libevent-2.1.so.7 => /usr/lib/x86_64-linux-gnu/libevent-2.1.so.7 (0x000014d19da21000)
			libdl.so.2 => /usr/lib/x86_64-linux-gnu/libdl.so.2 (0x000014d19da1b000)
			libutil.so.1 => /usr/lib/x86_64-linux-gnu/libutil.so.1 (0x000014d19da14000)
			libevent_pthreads-2.1.so.7 => /usr/lib/x86_64-linux-gnu/libevent_pthreads-2.1.so.7 (0x000014d19da0f000)
			libudev.so.1 => /usr/lib/x86_64-linux-gnu/libudev.so.1 (0x000014d19d9e3000)
			libltdl.so.7 => /usr/lib/x86_64-linux-gnu/libltdl.so.7 (0x000014d19d9d8000)

	Running the executable on two nodes, with four tasks per node, can be done like this:		Running the executable on two nodes, with four tasks per node, can be done like this:
	~~<source lang='bash'>~~		srun --nodes=2 --ntasks-per-node=4 --mpi=openmpi ./test_hello_world
	srun --nodes=2 --ntasks-per-node=4 --mpi=openmpi ./test_hello_world
	~~</source>~~

	This will result in the following output:		This will result in the following output:

Dawes0011 at 09:01, 16 June 2023

2023-06-16T09:01:59Z

← Older revision		Revision as of 11:01, 16 June 2023
Line 19:		Line 19:

	Before compiling, make sure that the compilers that are required available.		Before compiling, make sure that the compilers that are required available.
	<~~source lang='bash'~~>		<pre>
	module list		module list
	</~~source~~>		</pre>

	To avoid conflicts between libraries, the safest way is purging all modules:		To avoid conflicts between libraries, the safest way is purging all modules:

Dawes0011 at 09:01, 16 June 2023

2023-06-16T09:01:42Z

← Older revision		Revision as of 11:01, 16 June 2023
Line 44:		Line 44:

	linux-vdso.so.1 => (0x00002aaaaaacb000)		linux-vdso.so.1 => (0x00002aaaaaacb000)
	libmpi.so.1 => ~~/cm~~/shared/apps/openmpi/gcc/64/1.6.5/lib64/libmpi.so.1 (0x00002aaaaaccd000)		libmpi.so.1 => /shared/apps/openmpi/gcc/64/1.6.5/lib64/libmpi.so.1 (0x00002aaaaaccd000)
	libdl.so.2 => /lib64/libdl.so.2 (0x00002aaaab080000)		libdl.so.2 => /lib64/libdl.so.2 (0x00002aaaab080000)
	libm.so.6 => /lib64/libm.so.6 (0x00002aaaab284000)		libm.so.6 => /lib64/libm.so.6 (0x00002aaaab284000)

Dawes001 at 15:02, 15 July 2019

2019-07-15T15:02:51Z

← Older revision		Revision as of 17:02, 15 July 2019
Line 57:		Line 57:
	Running the executable on two nodes, with four tasks per node, can be done like this:		Running the executable on two nodes, with four tasks per node, can be done like this:
	<source lang='bash'>		<source lang='bash'>
	srun --nodes=2 --ntasks-per-node=4 ~~--partition=ABGC~~ --mpi=openmpi ./test_hello_world		srun --nodes=2 --ntasks-per-node=4 --mpi=openmpi ./test_hello_world
	</source>		</source>

Dawes001 at 15:02, 15 July 2019

2019-07-15T15:02:32Z

← Older revision		Revision as of 17:02, 15 July 2019
Line 77:		Line 77:
	#!/bin/sh		#!/bin/sh
	#SBATCH --comment=projectx		#SBATCH --comment=projectx
	#SBATCH --time=0		#SBATCH --time=30-0
	#SBATCH -n 32		#SBATCH -n 32
	#SBATCH --constraint=~~normalmem~~		#SBATCH --constraint=4gpercpu
	#SBATCH --output=output_%j.txt		#SBATCH --output=output_%j.txt
	#SBATCH --error=error_output_%j.txt		#SBATCH --error=error_output_%j.txt
	#SBATCH --job-name=MPItest		#SBATCH --job-name=MPItest
	~~#SBATCH --partition=ABGC_Production~~
	#SBATCH --mail-type=ALL		#SBATCH --mail-type=ALL
	#SBATCH --mail-user=user@wur.nl		#SBATCH --mail-user=user@wur.nl

Dawes001 at 10:50, 8 May 2018

2018-05-08T10:50:48Z

← Older revision		Revision as of 12:50, 8 May 2018
Line 76:		Line 76:
	$ vim batch.sh		$ vim batch.sh
	#!/bin/sh		#!/bin/sh
	#SBATCH --~~account~~=projectx		#SBATCH --comment=projectx
	#SBATCH --time=0		#SBATCH --time=0
	#SBATCH -n 32		#SBATCH -n 32

Bohme001: /* A mvapich2 sbatch example */

2014-03-28T13:10:45Z

A mvapich2 sbatch example

← Older revision		Revision as of 15:10, 28 March 2014
Line 92:		Line 92:
	echo "Running on $SLURM_NPROCS processors."		echo "Running on $SLURM_NPROCS processors."
	echo "Current working directory is `pwd`"		echo "Current working directory is `pwd`"
	echo "Env var MPIR_CVAR_NEMESIS_TCP_NETWORK_IFACE is $MPIR_CVAR_NEMESIS_TCP_NETWORK_IFACE"		# echo "Env var MPIR_CVAR_NEMESIS_TCP_NETWORK_IFACE is $MPIR_CVAR_NEMESIS_TCP_NETWORK_IFACE"

	# export MPIR_CVAR_NEMESIS_TCP_NETWORK_IFACE=ib0		# export MPIR_CVAR_NEMESIS_TCP_NETWORK_IFACE=ib0

Bohme001: /* A mvapich2 sbatch example */

2014-03-28T13:10:18Z

A mvapich2 sbatch example

← Older revision		Revision as of 15:10, 28 March 2014
Line 71:		Line 71:

	== A mvapich2 sbatch example ==		== A mvapich2 sbatch example ==
	A mpi job using mvapich2 on 32 cores using the fast infiniband interconnect for RDMA traffic.		A mpi job using mvapich2 on 32 cores, using the normal compute nodes and the fast infiniband interconnect for RDMA traffic.
	<source lang='bash'>		<source lang='bash'>
	$ module load mvapich2/gcc		$ module load mvapich2/gcc
Line 99:		Line 99:

	echo "Program finished with exit code $? at: `date`"		echo "Program finished with exit code $? at: `date`"

			$ sbatch batch.sh

	</source>		</source>

Bohme001: /* A simple 'Hello World' example */

2014-03-28T13:08:02Z

A simple 'Hello World' example

← Older revision		Revision as of 15:08, 28 March 2014
Line 69:		Line 69:
	Hello MPI! Process 0 of 8 on node010		Hello MPI! Process 0 of 8 on node010
	Hello MPI! Process 3 of 8 on node010		Hello MPI! Process 3 of 8 on node010

			== A mvapich2 sbatch example ==
			A mpi job using mvapich2 on 32 cores using the fast infiniband interconnect for RDMA traffic.
			<source lang='bash'>
			$ module load mvapich2/gcc
			$ vim batch.sh
			#!/bin/sh
			#SBATCH --account=projectx
			#SBATCH --time=0
			#SBATCH -n 32
			#SBATCH --constraint=normalmem
			#SBATCH --output=output_%j.txt
			#SBATCH --error=error_output_%j.txt
			#SBATCH --job-name=MPItest
			#SBATCH --partition=ABGC_Production
			#SBATCH --mail-type=ALL
			#SBATCH --mail-user=user@wur.nl

			echo "Starting at `date`"
			echo "Running on hosts: $SLURM_NODELIST"
			echo "Running on $SLURM_NNODES nodes."
			echo "Running on $SLURM_NPROCS processors."
			echo "Current working directory is `pwd`"
			echo "Env var MPIR_CVAR_NEMESIS_TCP_NETWORK_IFACE is $MPIR_CVAR_NEMESIS_TCP_NETWORK_IFACE"

			# export MPIR_CVAR_NEMESIS_TCP_NETWORK_IFACE=ib0

			mpirun -iface ib0 -np 32 ./tmf_par.out -NX 480 -NY 240 -alpha 11 -chi 1.3 -psi_b 5e-2 -beta 0.0 -zeta 3.5 -kT 0.10

			echo "Program finished with exit code $? at: `date`"

			</source>

Megen002 at 22:40, 12 December 2013

2013-12-12T22:40:13Z

← Older revision		Revision as of 00:40, 13 December 2013
Line 1:		Line 1:

			== A simple 'Hello World' example ==
	Consider the following simple MPI version, in C, of the 'Hello World' example:		Consider the following simple MPI version, in C, of the 'Hello World' example: