CUDA

Full CUDA course — 42 topics from the basics to advanced use.

Lessons

• Performance Analysis
• Optimization Techniques
• Cooperative Groups
• Dynamic Parallelism
• CUDA Graphs
• Stream-Ordered Memory Allocation
• Mixed Precision & Tensor Cores
• Modern Architecture Features
• CUDA C++
• CUDA Libraries
• Thrust & CUB
• Multi-GPU Programming
• Debugging & Correctness
• PTX, SASS & NVRTC
• Interoperability
• Python & Other Languages
• Portability
• Introduction to GPU Computing & CUDA
• GPU Architecture
• Installation, Toolkit & nvcc
• Your First Kernel
• The Thread Hierarchy
• Memory Management Basics
• Error Handling
• Device Query & Compute Capability
• Application Case Studies
• Best Practices
• Capstone Projects
• Appendices
• The Memory Hierarchy
• Global Memory & Coalescing
• Shared Memory & Tiling
• Synchronization & Atomics
• Warp-Level Programming
• Reduction & Scan
• More Parallel Patterns
• Constant, Texture & Surface Memory
• Streams, Events & Asynchronous Execution
• Pinned Memory & Data Transfer
• Unified Memory
• Occupancy & Launch Configuration
• Course Overview