https://github.com/neoblizz/spmv

Efficient Sparse Matrix-Vector Multiplication (SpMV) using ModernGPU (MTX + CSR formats).
https://github.com/neoblizz/spmv

csr cuda gpgpu load-balancing mtx spmv

Last synced: 3 days ago
JSON representation

Efficient Sparse Matrix-Vector Multiplication (SpMV) using ModernGPU (MTX + CSR formats).

• Host: GitHub
• URL: https://github.com/neoblizz/spmv
• Owner: neoblizz
• Created: 2020-08-08T17:25:24.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2021-10-22T16:05:36.000Z (over 3 years ago)
• Last Synced: 2024-12-17T15:46:45.589Z (about 2 months ago)
• Topics: csr, cuda, gpgpu, load-balancing, mtx, spmv
• Language: C
• Homepage:
• Size: 126 KB
• Stars: 2
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Load-Balanced Sparse Matrix-Vector Multiplication (SpMV)

Efficient Sparse Matrix-Vector Multiplication (SpMV) using ModernGPU (MTX + CSR formats). 

For more details, please see [moderngpu](https://github.com/moderngpu/moderngpu/wiki/Load-balancing-search#sparse-matrix--vector) 

for more details. This repository is a simple wrapper around moderngpu's implementation with 

support for loading matrix market format (MTX) using compressed sparse row (CSR) format.