- Infos im HLRS Wiki sind nicht rechtsverbindlich und ohne Gewähr -
- Information contained in the HLRS Wiki is not legally binding and HLRS is not responsible for any damages that might result from its use -

NEC Cluster Using MPI: Difference between revisions

From HLRS Platforms
Jump to navigationJump to search
 
(10 intermediate revisions by 2 users not shown)
Line 1: Line 1:
=== OpenMPI example ===
=== OpenMPI ===


==== simple example ====
see [[Open MPI]]


To use OpenMPI with intel Compiler, create a .modulerc in your home
=== Intel MPI ===
with this contents:


#%Module1.0#
see [[Intel MPI]]
set version 1.0
module load compiler/intel/11.0
module load mpi/openmpi/1.3-intel-11.0


For compilation use the mpi wrapper scripts like mpicc/mpic++/mpif90.
=== MVAPICH2 ===


The following example is for a pure MPI job, using 16 nodes (128 processes).
see [[MVAPICH2]]
For Illustration, this is done using an interactvie session (-I option)


First step: Batch submit to get the nodes
=== MPI I/O ===


  qsub -l nodes=16:nehalem:ppn=8,walltime=6:00:00 -I            # get the 16 nodes
see [[MPI-IO]]
 
In the session you will get after some time, the application is started with
mpirun -np 128 PathToYourApp
 
==== more complex examples ====
 
OpenMPI the resources in something called 'slots'.
By specifying 'ppn:X' to the batchsystem, the number of slots per node is specified.
So for a simple MPI job with 8 process per node (=1 process per core) ppn:8
is best choice, as in above example. Details can be specified on mpirun commandline.
PBS setup is adjusted for ppn:8, please do not use other values.
 
If you want to use less processes per node e.g. because you are restricted by memory per process,
or you have a hybrid parallel application using OpenMP and MPI,
MPI would always put the first 8 processes on the first node, second 8 on second and so on.
To avoid this, you can do
 
mpirun -np X -npernode 2 /path/to/app
 
This would start 2 processes per node. Like this, you can use a larger number of nodes
with a smaller number of processes, or you can e.g. start threads out of the processes.
 
If you want to pin your processes to a CPU (and enable NUMA memory affinity) use
 
 
mpirun -np X --mca mpi_paffinity_alone 1  /path/to/app
 
Warning: This will not behave as expected for hybrid multithreaded applications,
as the threads will be pinned to a single CPU as well! Use this only in case
of one process per core, no extra threads.
 
For pinning of hybrid OpenMP/MPI, you can use the wrapper from the intel MPI example,
and do not use mpi_paffinity_alone switch, but
 
mpirun -np X -npernode 2 /path/to/wrapper.sh /path/to/app
 
 
 
 
 
 
=== Intel MPI example ===
 
==== simple example ====
 
First load the neccessary modules
 
module load mpi/impi/intel-11.0.074-impi-3.2.0.011
 
Run your application with
mpirun -r ssh -np 8 app
 
 
==== more complex example ====
 
As Nehalem system is a two socket system with local attached ccNUMA memory,
memory and process placement can be crucial.
 
Here is an example that shows a 16 node Job, using 1 process per socket and 4 threads
per socket and optimum NUMA placement of processes and memory.
 
Prerequisite: Use intel MPI and best intel compiler
To setup environment for this, use this .modulerc file in your home:
 
{{File
| filename = .modulerc
| content = <pre>
#%Module1.0#
set version 1.0
module load compiler/intel/11.0
module load mpi/impi/intel-11.0.074-impi-3.2.0.011
</pre>
}}
 
And compile your application using mpicc/mpicxx/mpif90 (GNU compiler) or mpiicc/mpiicpc/mpiifort (Intel compiler).
 
First step: Batch submit to get the nodes
 
{{Command
| command = qsub -l nodes=16:nehalem:ppn=8,walltime=6:00:00 -I          # get the 16 nodes
}}
 
Second step: make a hostlist
{{Command
| command = sort -u  $PBS_NODEFILE  > m
}}
 
Third step: make a process ring to be used by MPI later
{{Command
| command = mpdboot  -n 16 -f m -r ssh 
}}
 
Fourth step: start MPI application
{{Command
| command = mpiexec -perhost 2 -genv I_MPI_PIN 0  -np 32 ./wrapper.sh ./yourGloriousApp
}}
 
With wrapper.sh looking like this
 
{{File
| filename = wrapper.sh
| content =<pre>
#!/bin/bash
export KMP_AFFINITY=verbose,scatter
export OMP_NUM_THREADS=4
RANK=${OMPI_COMM_WORLD_RANK:=$PMI_RANK}
if [ $(expr $RANK % 2) = 0  ]
then
    export GOMP_CPU_AFFINITY=0-3
    numactl --preferred=0 --cpunodebind=0 $@
else
    export GOMP_CPU_AFFINITY=4-7
    numactl --preferred=1 --cpunodebind=1 $@
fi
</pre>
}}
 
Result is an application running on 16 nodes, using 32 processes spawning
128 threads. One set of 4 threads is pinned to the one socket, the other set of 4 threads to the other socket.
 
=== MVAPICH2 example ===
 
==== simple example ====
 
To use MVAPICH2 first load the necessary module
 
module load mpi/mvapich2
 
You run your application with
 
mpirun_rsh -np 8 -hostfile $PBS_NODEFILE app

Latest revision as of 14:44, 12 June 2013

OpenMPI

see Open MPI

Intel MPI

see Intel MPI

MVAPICH2

see MVAPICH2

MPI I/O

see MPI-IO