RegDem: Increasing GPU Performance via Shared Memory Register Spilling.
Putt SakdhnagoolAmit SabneRudolf EigenmannPublished in: CoRR (2019)
Keyphrases
- shared memory
- parallel computing
- parallel programming
- parallel computation
- message passing
- graphic processing unit
- compute unified device architecture
- parallel algorithm
- parallel architectures
- multi processor
- parallel architecture
- distributed memory
- commodity hardware
- parallel execution
- shared memory multiprocessors
- multithreading
- parallel computers
- parallel machines
- parallel processing
- parallel implementation
- massively parallel
- single processor
- shared memory multiprocessor
- computer architecture
- graphics processing units
- memory access
- processing units
- high performance computing
- graph cuts