Benchmarking the NVIDIA V100 GPU and Tensor Cores.
Matt MartineauPatrick AtkinsonSimon McIntosh-SmithPublished in: Euro-Par Workshops (2018)
Keyphrases
- general purpose computing
- graphics processing units
- graphics hardware
- graphics processors
- scientific computing
- parallel implementation
- gpu implementation
- computing systems
- general purpose
- parallel computing
- compute unified device architecture
- high order
- parallel architectures
- cpu implementation
- higher order
- real time
- parallel processing
- highly parallel
- parallel computation
- high performance computing
- high end
- tensor space
- diffusion tensor
- parallel algorithm
- times faster
- dimensionality reduction
- tensor analysis
- massively parallel
- structure tensor
- tensor decomposition
- processing speed
- tensor factorization
- level parallelism