MPI_Dims_create - Creates a division of processors in a
Cartesian grid
SYNOPSIS
C:
#include <mpi.h>
int MPI_Dims_create(nnodes, ndims, dims)
int nnodes;
int ndims;
int *dims;
C++:
#include <mpi.h>
void Compute_dims(
int nnodes,
int ndims,
int dims[])
Fortran:
INCLUDE "mpif.h" (or USE MPI)
INTEGER nnodes, ndims, dims(*), ierror
CALL MPI_DIMS_CREATE(nnodes, ndims, dims(*),
ierror)
STANDARDS
This release implements the MPI 1.2 standard, as
documented by the MPI Forum in the spring 1997 release of
MPI: A Message Passing Interface Standard.
DESCRIPTION
The MPI_Dims_create routine creates a division of
processors in a Cartesian grid. This routine accepts the
following parameters:
nnodes Specifies the number of nodes in a grid
(integer)
ndims Specifies the number of Cartesian dimensions
(integer)
dims Returns an integer array of size ndims that
specifies the number of nodes in each dimension
completion, which is in MPI_SUCCESS.
MPI_SUCCESS is defined in the mpif.h file.
Man(1) output converted with
man2html