CUDA
1. Introduction
Code
- None
Exercise
- None
2. Data Parallel Computing
Code
- vector_add
- use __global__ methods
- slow add
- vector_add_kernel
- fast add
- use multi threads simultaneously
Exercise
3. Scalable Parallel Execution
Code
- grey
- control 2D images
- blur
- implement convolution filter
- fractal
- 3 dimensions kernel
- implement fractal
- unified memory
Exercise