Evaluating the Performance of the hipSYCL Toolchain for HPC Kernels on NVIDIA V100 GPUs.
Brian HomerdingJohn R. TrammPublished in: IWOCL (2020)
Keyphrases
- graphics processing units
- high performance computing
- graphics hardware
- general purpose
- gpu implementation
- kernel function
- parallel computing
- computing systems
- graphics processors
- massively parallel
- floating point
- parallel processing
- kernel methods
- linear combination
- parallel computation
- support vector
- parallel programming
- fault tolerance
- feature space
- compute unified device architecture
- parallel architectures
- cpu implementation
- computational power
- parallel implementation
- gaussian kernel
- scientific computing
- grid computing
- neural network
- data sets