GPU Programming
CS 101c
The use of Graphics Processing Units for rendering is well known, but their power for general parallel computation has only recently been explored. Parallel algorithms running on GPUs can often achieve up to 100x speedup over similar CPU algorithms, with many existing applications for physics simulations, signal processing, financial modeling, neural networks, and countless other fields.
This course will cover programming techniques for the GPU, focusing on visualization and simulation of various systems. Labs will cover specific applications in graphics, physics, and signal processing. The course will introduce the OpenGL Shader Language (GLSL) and NVidia's new parallel computing architecture, CUDA.
Labwork will require extensive programming. Some experience with computer graphics algorithms is preferred, but not required. A working knowledge of the C programming language will be necessary.9 units; third term.
Russell McClellan - russellm@caltech.edu
Tamas Szalay - tamas@caltech.edu
Professor Al Barr - barradmin@cs.caltech.edu
Jorgensen 74
Tuesday: Luke Durant, 8pm - 9pm
Russell McClellan, 9pm - 10pm
JRG CS Lab -- (For now)
100% Assignments, 60% to pass, but must make a reasonable attempt at all assignments. Assignments due Wednesdays at 1pm.
- GLSL Language and Syntax
- Phong shading, texture mapping on gpu
- Normal-mapping
- Vertex shaders
- Simulation using GLSL shaders
- CUDA architecture
- GPU hardware architecture
- CUDA syntax
- OpenGL/CUDA interoperability
- Parallel algorithm basics
- GPU numerics
- CUDA optimization