Heterogeneous CPU+GPU parallelization for high-accuracy scale-resolving simulations of compressible turbulent flows on hybrid supercomputers.
Andrey V. GorobetsPavel A. BakhvalovPublished in: Comput. Phys. Commun. (2022)
Keyphrases
- high accuracy
- graphics processing units
- shared memory
- heterogeneous computing
- parallel computing
- parallel processing
- compute unified device architecture
- graphic processing unit
- graphics processors
- high performance computing
- massively parallel
- distributed memory
- parallel computation
- parallel programming
- multithreading
- gpu implementation
- general purpose
- parallel execution
- commodity hardware
- scale space
- parallel implementation
- real time
- high efficiency
- parallel architectures
- heterogeneous systems
- message passing interface
- compute intensive
- data transfer
- message passing
- parallel algorithm
- memory bandwidth
- compressive sensing
- computing systems
- numerical simulations
- multiscale
- gpu accelerated
- fluid flow
- processing units
- highly accurate