Automatic Generation of Micro-kernels for Performance Portability of Matrix Multiplication on RISC-V Vector Processors.
Francisco D. IgualLuis PiñuelSandra CatalánHéctor MartínezAdrián CastellóEnrique S. Quintana-OrtíPublished in: SC Workshops (2023)
Keyphrases
- matrix multiplication
- distributed memory
- instruction set
- message passing
- shared memory
- dot product
- application specific
- parallel algorithm
- automatically generate
- kernel function
- feature vectors
- parallel implementation
- feature space
- matrix factorization
- kernel methods
- fisher kernel
- parallel computing
- floating point
- parallel processing
- rows and columns
- support vector
- probabilistic model