Evaluating the Performance of NVIDIA's A100 Ampere GPU for Sparse and Batched Computations.
Hartwig AnztYuhsiang M. TsaiAhmad AbdelfattahTerry CojeanJack J. DongarraPublished in: PMBS@SC (2020)
Keyphrases
- parallel implementation
- graphics processing units
- graphics processors
- graphics hardware
- gpu implementation
- parallel computation
- general purpose
- real time
- quality assurance
- high dimensional
- sparse data
- parallel computing
- parallel processing
- sparse coding
- sparse representation
- cpu implementation
- dictionary learning
- compressed sensing
- compute unified device architecture
- high performance computing
- floating point
- graphical models
- software engineering
- feature selection
- neural network