A Study of SpMV Implementation Using MPI and OpenMP on Intel Many-Core Architecture

@inproceedings{Ye2014ASO,
  title={A Study of SpMV Implementation Using MPI and OpenMP on Intel Many-Core Architecture},
  author={Fan Ye and Christophe Calvin and Serge G. Petiton},
  booktitle={VECPAR},
  year={2014}
}
The Sparse Matrix-Vector Multiplication is the key operation in many iterative methods. The widely used CSR (Compressed Sparse Row)[7] format was chosen to carry on this study for sustainability and reusability reasons. We parallelized for Intel MIC architecture a vectorized SpMV kernel using respectively the pure and the hybrid MPI/OpenMP models. In comparison to pure models and vendorsupplied BLAS libraries across different mainstream architectures (CPU, GPU), the hybrid model exhibits a… CONTINUE READING

References

Publications referenced by this paper.

Similar Papers

Loading similar papers…