Parallel and distributed computing are techniques to improve speed and handle large-scale tasks by dividing problems into smaller parts. Parallel computing uses multiple processors in one system (shared memory) for simultaneous computation, while distributed computing connects multiple independent computers (network) to act as one system.

CUDA C implementation of the Lax-Wendroff scheme for solving the Ideal MHD equations in 3D

Implementation of One Sided Jacobi SVD using CUDA on Jetson TK1 embedded GPU

A C++ header-only library for parallel linear algebra on GPUs (CUDA/cuBLAS under the hood)

Parallelizing matrix multiplication using Cuda C. Tiling is also implemented to compare results. This repository is submitted as the third assignment for the CSC447 (Parallel Programming for Multicore and Cluster Systems) course at the Lebanese American University.

This team project is presented as the final project for the CSC447 (Parallel Programming for Multicore and Cluster Systems) course at the Lebanese American University under the supervision of Dr. Hamdan Abdellatef.

CLE Third Assignment - The objective of this project was to take the second general problem, which have been discussed in the lab classes and for which we have developed both a multithreaded and a multiprocess solution. The aim now was to convert it into a CUDA program to be ran in a GPU under Linux.

Reinforcement learning console based PacMan Game

Learning CUDA Programming

Co-occurrence matrices act as the input to many unsupervised learning algorithms, including those that learn word embedding, and modern spectral topic models. However, the computation of these inputs often takes longer time than the inference. While much thought has been given to implementing fast learning algorithms. The co-occurrence matrix co…

CPU-GPU Benchmarking

Cycle-accurate UMA fault latency and bandwidth measurement for NVIDIA GPUs. C and PTX. No Python. Pascal (SM 6.0) through Blackwell GB10 (SM 12.1).