AI032
Programming Massively Parallel Processors: A Hands-on Approach
The Evolution and Future of GPU Computing
Learning Objectives
- Trace the architectural shift from fixed-pipeline rasterizers to programmable unified shaders
- Identify key milestones in the development of GPGPU and programming models like CUDA
- Evaluate the impact of specialized hardware like Tensor Cores and Ray Tracing units
- Analyze future trends in heterogeneous computing, chiplets, and AI-optimized silicon