High-Performance Computing

Educational implementations of parallel and distributed computing with MPI, OpenMP, and CUDA.

Root directory: MPI basics, collective operations, and serial/parallel algorithms (dot product, matrix operations, numerical methods)
openmp/: Shared memory parallelism (parallel loops, reductions, iterative solvers)
mpi/: Distributed memory examples
assignment/: Advanced implementations (sorting, 1D/2D heat diffusion)
hpc_project/: GPU acceleration with CUDA (multi-GPU matrix-vector multiplication)

Compilation

MPI:

mpicc -o program program.c
mpirun -np <num_processes> ./program

OpenMP:

gcc -fopenmp -o program program.c
./program

CUDA:

nvcc -o program program.cu
./program

See notes.md for additional technical notes and hpc_project/project_data.md for performance benchmarks.

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
assignment		assignment
cuda		cuda
hpc_project		hpc_project
mpi/practice		mpi/practice
openmp		openmp
practice		practice
.gitignore		.gitignore
README.md		README.md
average.c		average.c
avg_sd.c		avg_sd.c
bcast.c		bcast.c
dot_product.c		dot_product.c
gather.c		gather.c
greetings.c		greetings.c
heat2d.c		heat2d.c
hello_world.c		hello_world.c
matrix_matrix_prod.c		matrix_matrix_prod.c
matrix_vec_prod.c		matrix_vec_prod.c
min_max_sum.c		min_max_sum.c
notes.md		notes.md
parallel_gauss_jacobi.c		parallel_gauss_jacobi.c
parallel_gauss_jacobi_gen.c		parallel_gauss_jacobi_gen.c
parallel_sum.c		parallel_sum.c
ping_pong.c		ping_pong.c
ping_pong_2.c		ping_pong_2.c
ping_pong_nb.c		ping_pong_nb.c
point2point.c		point2point.c
scatter.c		scatter.c
sendrecv.c		sendrecv.c
trapezoidal.c		trapezoidal.c
trapezoidal_better.c		trapezoidal_better.c