Open Source Your Knowledge, Become a Contributor

Technology knowledge has to be shared and made accessible for free. Join the movement.

Create Content

Let's Start to REALLY have fun MPI

With high probability, the first program ever that many of us saw was (some variant of) the “hello, world” program in Kernighan and Ritchie’s classic text.

#include <stdio.h> 
int main(void) {
	printf("hello, world\n"); 
	return 0;

Let’s write a program similar to the classic “hello, world” that makes some use of MPI. Instead of having each process simply printing a message, we’ll designate one process to handle the output: the other processes will send it their messages, and then it will print them.

In parallel programming, it’s common (one might say standard) for the processes to be identified by nonnegative integer ranks. So if there are p processes, the processes will have ranks 0,1,2,..., p−1. For our parallel “hello, world,” let’s make process 0 the designated process, whom the other processes will send messages.

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

int main(int argc, char** argv) {
    // Initialize the MPI environment
    MPI_Init(NULL, NULL);

    // Get the number of processes
    int world_size;
    MPI_Comm_size(MPI_COMM_WORLD, &world_size);

    // Get the rank of the process
    int world_rank;
    MPI_Comm_rank(MPI_COMM_WORLD, &world_rank);

    // Get the name of the processor
    char processor_name[MPI_MAX_PROCESSOR_NAME];
    int name_len;
    MPI_Get_processor_name(processor_name, &name_len);

    // Print off a hello world message
    printf("Hello world from processor %s, rank %d out of %d processors\n",
           processor_name, world_rank, world_size);

    // Finalize the MPI environment.

Compilation and execution

The details about how compiling and running the program depend on your system, so you may need to check with a local expert. In this coursebook, we’ll assume that you’re using a text editor to write the program source and the command line to compile and run it. Many systems use a command called mpicc for compilation:

$ mpicc −o mpi hello.c

Typically, mpicc is a script wrapper for the C compiler. A wrapper script is a script whose main purpose is to run some program. In this case, the program is the C compiler. The wrapper simplifies how to run the compiler by telling it where to find the necessary header files and which libraries to link with the object file.

Many systems also support program startup with mpiexec or mpirun:

$ mpirun −np <number of processes> mpi

So, to run the program with one process, we’d type

$ mpirun −np 1 mpi
Hello World

and to run the program with four processes, we’d type

$ mpirun −np 4 mpi

The following example uses 4 processes.

Hello World

How do we get from invoking mpirun to one or more lines of greetings? The mpirun command tells the system to start instances of our program. It may also tell the system which core should run each instance of the program. After the processes are running, the MPI implementation takes care of making sure that the processes can communicate with each other.

How to use more MPI processes than your physical cores

If you are running MPI program on your local machine, with a particular number of cores, the latest version of OpenMPI will check the number of cores available, and does not allow you to run MPI program by using a number of core greater than the available physical cores.

In order to avoid that, and test your program using more MPI processes than your available cores, you can run the mpirun command using the option --oversubscribe.

$ mpirun −np 10 mpi --oversubscribe
Open Source Your Knowledge: become a Contributor and help others learn. Create New Content