Extending abstract GPU APIs to shared memory.
Ferosh JacobPublished in: SPLASH/OOPSLA Companion (2010)
Keyphrases
- shared memory
- parallel computing
- parallel computation
- parallel programming
- graphic processing unit
- parallel algorithm
- compute unified device architecture
- message passing
- parallel architectures
- distributed memory
- multi processor
- parallel architecture
- parallel machines
- commodity hardware
- parallel computers
- massively parallel
- multithreading
- graphics processing units
- parallel execution
- parallel processing
- address space
- high performance computing
- memory access
- real time
- multi core processors
- computer architecture
- source code
- open source
- shared memory multiprocessors
- heterogeneous platforms
- multi core systems
- interprocess communication
- gpu implementation
- high quality