Performance evaluation of unified memory and dynamic parallelism for selected parallel CUDA applications.
Lukasz JarzabekPawel CzarnulPublished in: J. Supercomput. (2017)
Keyphrases
- shared memory
- parallel computing
- parallel computation
- parallel implementation
- parallel processing
- level parallelism
- parallel programming
- parallel execution
- massively parallel
- dynamic environments
- parallel computers
- data parallelism
- message passing
- processing elements
- main memory
- parallel algorithm
- parallel architectures
- multithreading
- computational power
- distributed memory
- memory bandwidth
- compute unified device architecture
- coarse grain
- fine grain
- memory access
- multicore processors
- compute intensive
- parallel hardware
- single instruction multiple data
- mobile robot
- memory footprint
- multi threaded
- memory management
- data partitioning
- limited memory
- distributed systems
- general purpose