- Infos im HLRS Wiki sind nicht rechtsverbindlich und ohne Gewähr -

Difference between revisions of "NEC Cluster Using MPI"

From HLRS Platforms
Jump to navigationJump to search
(16 intermediate revisions by 3 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:
see [[Intel MPI]]
set version 1.0
module load compiler/intel/11.0
module load mpi/openmpi/1.3-intel-11.0
For compilationuse 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.
== Intel MPI example ==
As Nehalem system is a two socket system with local attached ccNUMA memory,
memory and process placmeent 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.
Prerequiste: Use intel MPI and best intel compiler
To setup environment for this, use this .modulerc file in your home:
set version 1.0
module load compiler/intel/11.0
module load mpi/impi/intel-11.0.074-impi-
And compile your application using mpicc/mpif90.
First step: Batch submit to get the nodes
  qsub -l nodes=16:nehalem:ppn=8,walltime=6:00:00 -I          # get the 16 nodes
Second step: make a hostlist
  sort -u  $PBS_NODEFILE  > m
Third step: make a process ring to be used by MPI later
mpdboot  -n 16 -f m -r ssh 
Fourth step: start MPI application
mpiexec -perhost 2 -genv I_MPI_PIN 0  -np 32 ./wrapper.sh ./yourGloriousApp
With wrapper.sh looking like this
export KMP_AFFINITY=verbose,scatter
if [ $(expr $PMI_RANK % 2) = 0  ]
        export GOMP_CPU_AFFINITY=0-3
        numactl --preferred=0 --cpunodebind=0 $@
        export GOMP_CPU_AFFINITY=4-7
        numactl --preferred=1 --cpunodebind=1 $@
Result is an application running on 16 nodes, using 32 processes spawning
128 threads. One set of 4 therads is pinned to the one socket, the other set of 4 threads to the other socket.

Latest revision as of 14:44, 12 June 2013


see Open MPI

Intel MPI

see Intel MPI




see MPI-IO