Automatic CPU/GPU Generation of Multi-versioned OpenCL Kernels for C++ Scientific Applications.
Rafael SotomayorLuis Miguel SánchezFrancisco Javier García BlasJavier FernándezJosé Daniel GarcíaPublished in: Int. J. Parallel Program. (2017)
Keyphrases
- graphics processing units
- general purpose
- gpu implementation
- parallel computing
- graphics hardware
- graphics processors
- parallel processing
- fully automatic
- parallel programming
- computing systems
- real time
- massively parallel
- semi automatic
- efficient implementation
- parallel algorithm
- floating point
- kernel function
- compute unified device architecture
- multithreading
- parallel execution
- support vector
- parallel computation
- generation process
- parallel implementation
- scientific data
- kernel methods
- high performance computing
- multiple kernel learning
- feature space
- cpu implementation
- graphic processing unit