User Tools

Site Tools


MPI Scatter and Gather

MPI program that combines both MPI_Scatter and MPI_Gather.
In this example, the master process (rank 0) will scatter an array to all processes. Each process will perform some operation on its received portion, and then MPI_Gather will be used to collect these modified segments back into an array in the master process.

#include <mpi.h>
#include <stdio.h>
#include <stdlib.h>
int main(int argc, char** argv) {
    // Initialize the MPI environment
    MPI_Init(&argc, &argv);
    // Get the rank of the process
    int world_rank;
    MPI_Comm_rank(MPI_COMM_WORLD, &world_rank);
    // Get the total number of processes
    int world_size;
    MPI_Comm_size(MPI_COMM_WORLD, &world_size);
    // The master process (rank 0) prepares the data
    const int elements_per_proc = 3;
    int *scatter_data = NULL;
    int *gather_data = NULL;
    if (world_rank == 0) {
        const int total_elements = elements_per_proc * world_size;
        scatter_data = (int *)malloc(total_elements * sizeof(int));
        gather_data = (int *)malloc(total_elements * sizeof(int));
        // Initialize the array with values
        for (int i = 0; i < total_elements; i++) {
            scatter_data[i] = i + 1;
    // Each process will receive a portion of the array from the master process
    int recv_data[elements_per_proc];
    // Scatter the data from the master process to all processes
    MPI_Scatter(scatter_data, elements_per_proc, MPI_INT, recv_data, elements_per_proc, MPI_INT, 0, MPI_COMM_WORLD);
    // Each process performs some operation on the received data
    for (int i = 0; i < elements_per_proc; i++) {
        recv_data[i] *= 2; // Example operation: multiply each element by 2
    // Gather the modified data from all processes back to the master process
    MPI_Gather(recv_data, elements_per_proc, MPI_INT, gather_data, elements_per_proc, MPI_INT, 0, MPI_COMM_WORLD);
    // The master process prints the gathered data
    if (world_rank == 0) {
        printf("Gathered data: ");
        for (int i = 0; i < elements_per_proc * world_size; i++) {
            printf("%d ", gather_data[i]);
    // Clean up
    if (world_rank == 0) {
    // Finalize the MPI environment

Initialization: The MPI environment is initialized.

Data Preparation: The master process creates an array (scatter_data) and another array (gather_data) for the gathered results.

Scatter Operation: MPI_Scatter is used to distribute segments of the scatter_data array to each process.

Data Processing: Each process modifies its segment of data. In this example, each element is multiplied by 2.

Gather Operation: MPI_Gather collects the modified segments from each process and assembles them into the gather_data array in the master process.

Output: The master process prints the gathered data.

Cleanup: The master process frees the allocated memory.

Finalization: The MPI environment is finalized.

This program demonstrates how to use MPI_Scatter to distribute data and MPI_Gather to collect the results. The total number of elements in scatter_data should be a multiple of the number of processes.

c/mpi/scatter-gather.txt · Last modified: 2024/01/17 10:50 by odefta