A performance comparison of linear algebra libraries for sparse matrix-vector product
DOI:
https://doi.org/10.5540/03.2015.003.01.0116Keywords:
Sparse Matrix-vector product, Computational Linear Algebra, Linear Algebra librariesAbstract
In this work we evaluate the computing performance of some well-known linear algebra libraries when performing the sparse matrix-vector product using different matrix formats. We use as benchmark real-world sparse matrices that came from different domains. The results are carried out and compared on CPU and GPU processors.
Downloads
Download data is not yet available.
Downloads
Published
2015-08-25
Issue
Section
Computação Científica