Demystifying the Placement Policies of the NVIDIA GPU Thread Block Scheduler for Concurrent Kernels.
Guin GilmanSamuel S. OgdenTian GuoRobert J. WallsPublished in: SIGMETRICS Perform. Evaluation Rev. (2020)
Keyphrases
- graphics processing units
- graphics processors
- parallel implementation
- graphics hardware
- gpu implementation
- hierarchical reinforcement learning
- real time
- general purpose
- cpu implementation
- parallel computing
- compute unified device architecture
- parallel computation
- parallel processing
- instruction scheduling
- multiple kernel learning
- scheduling policies
- kernel function
- linear combination
- feature space
- temporally extended
- kernel methods
- times faster
- optimal policy
- kernel learning
- support vector
- scheduling algorithm
- mutual exclusion
- massively parallel
- computing systems
- efficient implementation
- reproducing kernel hilbert space
- dct coefficients
- floating point
- real time rendering
- shared memory
- image quality