Hybrid MPI and CUDA paralleled finite volume unstructured CFD simulations on a multi-GPU system.
Xi ZhangXiaohu GuoYue WengXianwei ZhangYutong LuZhong ZhaoPublished in: Future Gener. Comput. Syst. (2023)
Keyphrases
- parallel implementation
- parallel computing
- parallel programming
- parallel computation
- gpu implementation
- message passing interface
- shared memory
- gpu accelerated
- parallel algorithm
- compute unified device architecture
- general purpose
- graphics processors
- real time
- parallel processing
- graphics hardware
- parallel computers
- ray casting
- computing systems
- finite number
- massively parallel
- graphics processing units
- computer architecture
- parallel architectures
- hybrid learning
- scheduling problem