Lockless Inc


MPI_Comm_splits - Splits a communicator


#include <mpi.h> int MPI_Comm_split(MPI_Comm comm , int split_key , int key , MPI_Comm *out );

#include <pmpi.h> int PMPI_Comm_split(MPI_Comm comm , int split_key , int key , MPI_Comm *out );

#include <mpi.h> MPI::Intercomm MPI::Intercomm::Split(int split_key , int key ) const; MPI::Intracomm MPI::Intracomm::Split(int split_key , int key ) const;

INCLUDE 'mpif.h' MPI_COMM_SPLIT(comm , split_key , key , out , ierr ) INTEGER comm , split_key , key , out , ierr


comm - communicator (handle)

split_key - key to split communicator by (integer)

key - key to order processes in new communicators (integer)


out - the new communicator that this process belongs to (handle).


The MPI_Comm_split() function is used to split a communicator comm into smaller sub-communicators. This is done by using split_key . Processes that pass the same split_key will become members of the same output communicator. However, if a process passes a split_key with the value MPI_UNDEFINED then MPI_COMM_NULL will be returned. Once the size and membership of the communicators are determined, then key is used to order the processes within them. The new communicator that the current process belongs to is returned in result .

In C++, the result communicator is passed as the return value.

Note that the communicator must be valid (not MPI_COMM_NULL). PMPI_Comm_split() is the profiling version of this function.


All MPI routines except for MPI_Wtime and MPI_Wtick return an error code. The the current MPI error handler is invoked if the return value is not MPI_SUCCESS. The default error handler aborts, but this may be changed with by using the MPI_Errhandler_set() function. The predefined error handler MPI_ERRORS_RETURN may be used to cause error values to be returned instead. Note that MPI does not guarentee that an MPI program can continue past an error. In this implementation, all errors except MPI_ERR_INTERN or MPI_ERR_OTHER should always be recoverable.

In C, the error code is passed as the return value. In FORTRAN, all functions have an parameter ierr which returns the error code. MPI C++ functions do not directly return an error code. However, C++ users may want to use the MPI::ERRORS_THROW_EXCEPTIONS handler. This will throw an MPI::Exception with the corresponding error code. To prevent exceptions from being raised from within C and Fortran code, they will see all error return values as MPI_ERR_PENDING when this handler is chosen. In this implementation, call MPI::throw_exception() to throw the correct exception if this occurs.

MPI_SUCCESS - No error;

MPI_ERR_PENDING - Pending exception;

MPI_ERR_COMM - Invalid communicator;

MPI_ERR_ARG - Invalid output pointer;

MPI_ERR_INTERN - Out of Memory. This may be fatal.


MPI_Comm_dup (3) MPI_Comm_create (3) MPI_Comm_free (3) MPI_Intercomm_create (3)

About Us Returns Policy Privacy Policy Send us Feedback
Company Info | Product Index | Category Index | Help | Terms of Use
Copyright © Lockless Inc All Rights Reserved.