MPI (Message Passing Interface) is a standard for writing message passing programs between parallel processes. It was developed in the late 1980s and early 1990s due to increasing computational needs. An MPI program typically initializes and finalizes the MPI environment, declares variables, includes MPI header files, contains parallel code using MPI calls, and terminates the environment before ending. Key MPI calls initialize and finalize the environment, determine the process rank and number of processes, and get the processor name.

1. Message Passing
Interface(MPI)
Himanshi Kathuria

2. Content
MPI
1. Definition
2. Reason For Using MPI
3. History And Evolution
4. General MPI Program Structure
5. Writing MPI PROGRAMS
6. Compiling MPI Programs
7. Running MPI Programs
8. Environment Management Routines

3. MPI Message Passing Interface
A means of machine communication
DEFINITION
• The Message passing interface Standard is a message passing library
industry standard for parallel programming based on the consensus of the
MPI forum which has over 40 participating organization, including vendors,
researchers software library developers, and users
• MPI provides widely used standard for writing message passing programs.

4. Motivation
Motivated by high computational complexity and memory requirements of
large applications.

5. HISTORY AND EVOLUTION
• Late 1980s: vendors had unique libraries
• 1989: ParallelVirtual Machine (PVM) developed at Oak Ridge National Lab
• 1992:Work on MPI standard begun
• 1994:Version 1.0 of MPI standard
• June 1995 MPI 1.1 published (clarifications)
• July 1996 MPI 1.2 published (clarifications)
• April 1995 MPI 2.0 committee formed
• July 1997 MPI 2.0 published

6. MPI CONCEPT
•MPI uses objects called communicators and groups to define which collection of
processes may communicate with each other.
•Most MPI routines require you to specify a communicator as an argument.
•Communicators and groups will be covered in more detail later. For now, simply use
MPI_COMM_WORLD whenever a communicator is required - it is the predefined
communicator that includes all of your MPI processes.

7. Rank:
• Within a communicator, every process has its own unique, integer identifier
assigned by the system when the process initializes.
• A rank is sometimes also called a "task ID".
• Ranks are contiguous and begin at zero.
• Used by the programmer to specify the source and destination of messages.
Size :
• Within a communicator, no of processes is defined as size.

8. General MPI Program Structure:
MPI include file
Initialize MPI Environment
Do work and make messaging passing calls
Terminate MPI Environment
Variables declaration
Parallel code begins

9. Include Files
#include <mpi.h>
• MPI header file
#include <stdio.h>
• Standard I/O header file
Variables
int main (int argc, char *argv[]) {
int i;
int id; /* Process rank */
int p; /* Number of processes */
void check_circuit (int, int);
• Include argc and argv: they are needed to initialize MPI
• One copy of every variable for each process running this program
argc=argument count ie. Gives the total number argument given
argv[]= pointer the argument value

10. Initialize MPI ENVIRONMENT
MPI_Init (&argc, &argv);
• First MPI function called by each process
• Not necessarily first executable statement
• Allows system to do any necessary setup
Parameters
argc [in] Pointer to the number of arguments .
argv [in] Pointer to the argument vector
Terminate MPI ENVIRONMENT
MPI_Finalize();
• Call after all other MPI library calls
• Allows system to free up MPI resources

11. WRITING MPI PROGRAMS
#include "mpi.h" // Gives basic MPI types, definitions
#include <stdio.h>
int main( argc, argv )
int argc;
char **argv;
{
MPI_Init( &argc, &argv ); // Starts MPI
::
Actual code including normal ‘C’ calls and MPI calls
::
MPI_Finalize(); // Ends MPI
return 0;
}

12. Compiling MPI Programs
Compile step:
mpicc -o <object file> <source code>.c ( By default the output file is a. out)
Compilation commands
• – mpicc -o hello_world hello-world.c
Running MPI Programs
Execution steps:
mpiexec/mpirun -np <i> <object file>
• To run hello_world on two machines:
mpirun -np 2 hello_world
Must specify full path of ‘executable’
• To know the commands executed by mpirun
mpirun –t
• To get all the mpirun options
mpirun –help

13. Environment Management Routines
MPI environment management routines are used for initializing and terminating the MPI
environment, querying the environment etc. Most of the commonly used ones are:
• MPI_Init
• MPI_Comm_size
• MPI_Comm_rank
• MPI_Get_processor_name
• MPI_Finalize

14. MPI_Comm_size
Determines the number of processes in the group associated with a communicator.
int MPI_Comm_size( MPI_Comm comm, int *size);
Parameters
• comm [in] communicator (handle)
• size [out] number of processes in the group of comm (integer).
MPI_Comm_size(MPI_COMM_WORLD, &size)
MPI_Comm_rank
Determines the rank of the calling process in the communicator.
int MPI_Comm_rank( MPI_Comm comm, int *rank);
Parameters
• comm [in] communicator (handle)
• rank [out] rank of the calling process in the group of comm(integer)
MPI_Comm_rank(MPI_COMM_WORLD, &rank);

15. MPI_Get_processor_name
Gets the name of the processor.
int MPI_Get_processor_name( char *name, int *resultlen );
Parameters
• name [out] A unique specifier for the actual node.This must be an array
of size at least MPI_MAX_PROCESSOR_NAME.
• resultlen [out] Length (in characters) of the name
MPI_Get_processor_name(&name, &resultlength)

16. THANKYOU