MPI Parallel Programs

This repository contains two MPI-based parallel C programs for High-Performance Computing (HPC) coursework:

1. Problem 1: Counting Primes
   Count the number of prime numbers in a user-specified range [x, y] using two variants:

   • Method (a): MPI_Bcast + MPI_Reduce
   • Method (b): MPI_Send + MPI_Recv

2. Problem 2: Matrix Summation
   Compute the element-wise sum of two N×M matrices A and B in parallel, using MPI_Scatter and MPI_Gather.

.
├── problem1/
│   ├── primes_bcast_reduce.c     # Solution using MPI_Bcast & MPI_Reduce
│   ├── primes_bcast_reduce       # Executable
│   ├── primes_send_recv.c        # Solution using MPI_Send & MPI_Recv
│   └── primes_send_recv          # Executable
│
├── problem2/
│   ├── matrix_sum_scatter_gather.c  # Solution using MPI_Scatter & MPI_Gather
│   └── matrix_sum_scatter_gather    # Executable
│
└── README.md                      # (you are here)

• MPI implementation (e.g., Open MPI or MPICH)
• C compiler (mpicc)
• Unix-like shell (Linux/macOS)

Count the prime numbers within a user-defined range using:

• Method A: MPI_Bcast and MPI_Reduce only.
• Method B: MPI_Send and MPI_Recv only.

# Compile and run the Bcast/Reduce version
cd 'Problem 1'
mpicc primes_bcast_reduce.c -o primes_bcast_reduce -lm
mpirun -np 4 ./primes_bcast_reduce

Sample Output:

Enter first Number: 1
Enter second Number: 16
The number of prime numbers between 1 and 16 is: 6

# Compile and run the Send/Recv version
mpicc primes_send_recv.c -o primes_send_recv -lm
mpirun -np 4 ./primes_send_recv

Sample Output:

Enter first Number: 1
Enter second Number: 16
Process 1 found 3 prime numbers between 1 and 5
Process 2 found 1 prime numbers between 6 and 10
Process 3 found 2 prime numbers between 11 and 16
Total prime numbers between 1 and 16 are: 6

Sum two user-defined matrices using MPI_Scatter and MPI_Gather.

cd ../'Problem 2'
mpicc matrix_sum_scatter_gather.c -o matrix_sum_scatter_gather
mpirun -np 4 ./matrix_sum_scatter_gather

Sample Output:

Enter the size of the matrices (n m): 2 2
Enter the elements of matrix A (4 values):
1 2 3 4
Enter the elements of matrix B (4 values):
1 2 3 4
The sum matrix C is:
2 4 
6 8

• All programs are written in C using MPI.
• Compilation uses -lm for linking the math library in Problem 1.
• Programs are tested using 4 MPI processes.
• Both problems assume the work divides evenly across processes.
• In Problem 2, n*m must be divisible by the number of processes. Otherwise, the program will abort with an error.
• Ensure your MPI environment variables (e.g., PATH and LD_LIBRARY_PATH) are properly set for your MPI installation.