This project is a part of my thesis focusing on researching and applying the general-purpose graphics processing unit (GPGPU) in high performance computing. In this project, I applied GPU Computing and the parallel programming model CUDA to solve the diffusion equation.

Cuda from Scratch

This repo contains some CUDA C++ code examples that demonstrate how to use GPUs for parallel computing. Covering topics such as dynamic parallelization, Optimization, ....etc

Implementation of a cellular automaton on GPU using different features of CUDA

Educational notes (for beginners) for parallel programming with CUDA (and code)

A demo code that compares the speed of two matrix multiplication functions executed by the CPU and the GPU

Codigo hecho en C/Cuda, tratamaos de procesar una imagen mediante una GPU de Nvidia. EL procesamiento de imagenes se trata de pasarles un filtro a la imagen proporcionada.

University project of the Parallel Computing course. There are two challenges that are solved using parallel computing techniques and two famous frameworks: OpenMP and CUDA.

Numerical integration (trapezoidal rule) with CUDA: performance benchmark.

Speed up the computation of local node connectivity with CUDA

This repository contains materials, notes, and projects for the High-Performance Computing (HPC) course in the Data Science degree at IPN. It covers the design and implementation of parallel computing, distributed systems, and GRID architectures to build high-performance computing systems.

GPU-accelerated Canny edge detector in CUDA C++. Parallelizes Gaussian filtering, gradient computation, non-maximum suppression, and hysteresis thresholding for real-time edge detection performance

Looping over SHA-256 hashes using parallelization on GPU. NVIDIA - Cuda

Chess engine for WUT GPU course

Parallel computing with CUDA to implement the heat spreading equation in 2 dimensions in the case of a metal plate starting with a circular hotspot in the middle of it. The programming language used for the visualization of the plate in different moments in time is Python.

Repo for CUDA C++ GPU kernels for ML and HPC.

Multi-Max-Clique, an application that solves Maximum Clique Problem using the parallel branch and bound approach and achieved linear and super-linear speedups in CUDA.

An implementation of Principal Direction Divisive Partitioning in CUDA. University project for the course "Software & Programming of High Perfomance Systems". Course Code: CEID_NE5407

A benchmark for dynamic-programming-based GPU sequence alignment algorithms.

A comparative study of high-performance neural network evaluation using OpenMP for multi-core CPUs and CUDA for NVIDIA GPUs. Includes performance analysis and speedup benchmarks.