On-the-fly workload partitioning for integrated CPU/GPU architectures.
Younghyun ChoFlorian NegeleSeohong ParkBernhard EggerThomas R. GrossPublished in: PACT (2018)
Keyphrases
- heterogeneous computing
- gpu implementation
- graphics processing units
- graphics processors
- real time
- neural network
- parallel architectures
- parallel computing
- data transfer
- graphics hardware
- personal computer
- response time
- high performance computing
- parallel implementation
- partitioning algorithm
- computer graphics
- memory hierarchy
- general purpose