Collectives

Reduction

C | Fortran-2008 | Fortran-90

MPI_Reduce

Definition

MPI_Reduce is the means by which MPI process can apply a reduction calculation. The values sent by the MPI processes will be combined using the reduction operation given and the result will be stored on the MPI process specified as root. MPI_Reduce is a collective operation; it must be called by every MPI process in the communicator given. Predefined operations are: MPI_MIN, MPI_MAX, MPI_BOR, MPI_BXOR, MPI_LOR, MPI_LXOR, MPI_BAND, MPI_LAND, MPI_SUM and MPI_PROD. Other variants of MPI_Reduce are MPI_Ireduce, MPI_Allreduce, MPI_Iallreduce. Refer to MPI_Ireduce to see the blocking counterpart of MPI_Reduce.

Copy

Feedback

int MPI_Reduce(const void* send_buffer,
               void* receive_buffer,
               int count,
               MPI_Datatype datatype,
               MPI_Op operation,
               int root,
               MPI_Comm communicator);

Parameters

send_buffer

A pointer on the buffer to send for reduction.

receive_buffer

A pointer on the buffer in which store the result of the reduction. Only the MPI process with the specified as root will receive the reduction result.

count

The number of elements in the send buffer, which is identical to that in the receive buffer as well.

datatype

The type of a buffer element.

operation

The operation to apply to combine messages received in the reduction. This operation must be associative, and commutative for predefined operations while user-defined operations may be non-commutative.

root

The rank of the MPI process that will collect the reduction result.

communicator

The communicator in which the reduction takes place.

Return value

The error code returned from the reduction.

Example

Copy

Feedback

#include <stdio.h>
#include <stdlib.h>
#include <mpi.h>

/**
 * @brief Illustrates how to use a reduce.
 * @details This application consists of a sum reduction; every MPI process
 * sends its rank for reduction before the sum of these ranks is stored in the
 * root MPI process. It can be visualised as follows, with MPI process 0 as
 * root:
 *
 * +-----------+ +-----------+ +-----------+ +-----------+
 * | Process 0 | | Process 1 | | Process 2 | | Process 3 |
 * +-+-------+-+ +-+-------+-+ +-+-------+-+ +-+-------+-+
 *   | Value |     | Value |     | Value |     | Value |
 *   |   0   |     |   1   |     |   2   |     |   3   |
 *   +-------+     +-------+     +-------+     +-------+
 *            \         |           |         /
 *             \        |           |        /
 *              \       |           |       /
 *               \      |           |      /
 *                +-----+-----+-----+-----+
 *                            |
 *                        +---+---+
 *                        |  SUM  |
 *                        +---+---+
 *                            |
 *                        +---+---+
 *                        |   6   |
 *                      +-+-------+-+
 *                      | Process 0 |
 *                      +-----------+
 **/
int main(int argc, char* argv[])
{
    MPI_Init(&argc, &argv);

    // Determine root's rank
    int root_rank = 0;

    // Get the size of the communicator
    int size = 0;
    MPI_Comm_size(MPI_COMM_WORLD, &size);
    if(size != 4)
    {
        printf("This application is meant to be run with 4 MPI processes.\n");
        MPI_Abort(MPI_COMM_WORLD, EXIT_FAILURE);
    }

    // Get my rank
    int my_rank;
    MPI_Comm_rank(MPI_COMM_WORLD, &my_rank);

    // Each MPI process sends its rank to reduction, root MPI process collects the result
    int reduction_result = 0;
    MPI_Reduce(&my_rank, &reduction_result, 1, MPI_INT, MPI_SUM, root_rank, MPI_COMM_WORLD);

    if(my_rank == root_rank)
    {
        printf("The sum of all ranks is %d.\n", reduction_result);
    }

    MPI_Finalize();

    return EXIT_SUCCESS;
}