Sparse matrix partitioning for optimizing SpMV on CPU-GPU heterogeneous platforms.
Akrem BenatiaWeixing JiYizhuo WangFeng ShiPublished in: Int. J. High Perform. Comput. Appl. (2020)
Keyphrases
- sparse matrix
- heterogeneous platforms
- graphics processing units
- graphics processors
- gpu implementation
- rows and columns
- floating point
- web services
- shared memory
- general purpose
- parallel implementation
- parallel computing
- random projections
- fixed point
- higher order
- pattern recognition
- graph partitioning
- mobile applications
- sensitivity analysis
- pairwise