Learning CUDA

This repo is for storing my CUDA learning notes and code.

Table of Contents

I have divided the learning process into serval units:

• Intro To GPU
• What is CUDA
• CUDA Programming Model
• CUDA Memory Model
• CUDA Thread Model
• CUDA Thread Synchronization
• CUDA Tools
• CUDA Libraries
• NVVC Compiler
• ncu tools

There is also a folder called general_knowledge where I will store general knowledge about parallel computing in general.

• Amdahl's Law
• SIMD
• SSE4

Code Examples

1. Hello World
2. Vector Addition
3. Dot Product
4. Matrix Multiplication
5. Tiled Matrix Multiplication
6. Matrix Transpose
7. CUDA Streams
8. CUDA Events
9. CUDA Atomics
10. cuBLAS
11. cuFFT
12. Threads Synchronization

Note

In this repo we don't tell you how to install or compile CUDA code. We are here to give the basic understanding of CUDA programming.

Learning Resources

Written

• CUDA Programming Guide Official Guide from NVIDIA.
• CUDA Book
• Repo for the CUDA Course

Video

• CUDA Tools
• CUDA Programming
• GPU Programming
• CUDA Course