- 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 -

MPT

From HLRS Platforms
Jump to navigationJump to search
MPT is the Message Passing Toolkit from HPE. It includes HPE's proprietary MPI implementation and is available on systems from HPE, e.g., HAWK.
Developer: HPE
Platforms: HAWK
Category: MPI
License: commercial, HPE
Website: HPE


Usage

This shows the basic steps to use the global MPT installations on the systems at HLRS.

Load the necessary module to set up the environment

module load mpt # on hawk


Then compile your application using the compiler wrappers mpicc, mpicxx or mpif08:

mpicc your_app.c - o your_app


The wrapper compilers will pass any necessary options, e.g. for include files or libraries to be linked to, to the underlying base compiler and linker (e.g. GNU, Intel, AOCC).

Now you can allocate compute nodes via the Batch_system and run your application with

mpirun <OPTIONS> your_app

The most important options you may need on our systems are described in the following sections. For in depth information of all options we refer you to the man pages of mpirun and MPI.

module load mpt

man mpirun # mpirun usage
man omplace # binding tool for OpenMP applications

man MPI # description of environment variables for MPT (advanced features)


Examples

For a collection of usage examples see Batch_System_PBSPro_(Hawk)

Common Problems

Running applications via wrapper scripts

Many tools run MPI applications through wrapper scripts, e.g., mpirun ./wrapper.sh original_mpi_program. With MPT this is only possible when using a separate shepherd. To enable this mode use

export MPI_SHEPHERD=true
mpirun ./wrapper.sh original_mpi_program


See also

External links