High-Performance Matrix-Vector Multiplication on the GPU.
Hans Henrik Brandenborg SørensenPublished in: Euro-Par Workshops (1) (2011)
Keyphrases
- graphics processing units
- sparse matrix
- floating point
- matrix representation
- transformation matrix
- eigenvalues and eigenvectors
- real time
- matrix multiplication
- gpu implementation
- parallel computing
- rows and columns
- linear algebra
- dot product
- weight matrix
- gpu accelerated
- fuzzy membership
- graphics hardware
- parallel implementation
- vector space
- cost effective
- symmetric matrices
- linearly independent
- general purpose
- pairwise
- modular exponentiation
- symmetric matrix
- gradient vector
- cluster of workstations
- coefficient matrix
- vector data
- highly parallel
- singular values
- distributed memory
- high efficiency
- sufficient conditions
- signal processing
- feature vectors