Implementing and Evaluating an Heterogeneous, Scalable, Tridiagonal Linear System Solver with OpenCL to Target FPGAs, GPUs, and CPUs.
Hamish J. MacintoshJasmine BanksNeil A. KelsonPublished in: Int. J. Reconfigurable Comput. (2019)
Keyphrases
- graphics processing units
- commodity hardware
- parallel algorithm
- parallel architectures
- parallel processing
- parallel programming
- general purpose
- linear systems
- shared memory
- parallel computing
- gpu implementation
- real time
- graphics hardware
- efficient implementation
- parallel computation
- field programmable gate array
- massively parallel
- parallel implementation
- graphics processors
- computing systems
- floating point
- processing units
- high end
- closed form
- lightweight
- multiple types
- high performance computing
- moving target
- linear model
- sufficient conditions
- memory bandwidth