https://github.com/sferez/sspp_sparse_matrix_cuda
Small Scale Parallel Programming, Sparse Matrix multiplication with CUDA
https://github.com/sferez/sspp_sparse_matrix_cuda
cpp cuda omp omp-parallel parallel-computing small-scale-parallel-programming sparse-matrix
Last synced: 9 months ago
JSON representation
Small Scale Parallel Programming, Sparse Matrix multiplication with CUDA
- Host: GitHub
- URL: https://github.com/sferez/sspp_sparse_matrix_cuda
- Owner: sferez
- Created: 2023-03-24T23:53:53.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2023-04-18T09:38:02.000Z (almost 3 years ago)
- Last Synced: 2025-03-03T03:42:05.980Z (about 1 year ago)
- Topics: cpp, cuda, omp, omp-parallel, parallel-computing, small-scale-parallel-programming, sparse-matrix
- Language: C++
- Homepage:
- Size: 66.5 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Small Scale Parallel Computing
This study examined the performance of OpenMP and CUDA parallelization for Sparse matrix-vector multiplication (SpMV) on a hybrid CPU and GPU platform.
It implemented both programming models on a set of sparse matrices with varying sizes and densities.
## Report
The report is available at Medium: [Article](https://medium.com/@simeon.ferez/sparse-matrix-vector-product-openmp-vs-cuda-on-hybrid-platforms-7ef576fbfc0d)
## Structure
The following code folders are included in this project:
- 'input' contains the input matrices in matrix market format
- 'output' contains the sparsity pattern of some output matrices
- 'src/CUDA' contains the CUDA code parallelization
- 'src/OMP' contains the OpenMP code parallelization
- 'src' contains the custom class and python script to run the code
## Requirements
- CMake
- CUDA
- OpenMP
- Python 3 (optional)
## Build
Each folder contains a CMakeLists.txt file to build the code. Each parallelization is built separately for testing purpose.
Python script have been provided to run all the configurations and generate the results.
## Authors
- [@sferez](https://github.com/sferez)