GPUs and CUDA¶

Getting started¶

• NVIDIA blog: An Even Easier Introduction to CUDA (Mark Harris)
• CUDA Refresher
• Getting Started with CUDA and Parallel Programming (video)

Understanding¶

• MolSSi - Fundamentals of Heterogeneous Parallel Programming with CUDA C/C++
• Cornell Virtual Workshop - Understanding GPU Architecture
• NVIDIA blog: How to Optimize Data Transfers in CUDA C/C++ (Mark Harris)
• NVIDIA blog: How to Overlap Data Transfers in CUDA C/C++ (Mark Harris)
• NVIDIA blog: How to Implement Performance Metrics in CUDA C/C++ (Mark Harris)
• Roofline performance model (NERSC)

Resources¶

• The Future Is Tiled: Using CuTile and TileIR To Write Portable, High-performance GPU Kernels (video)

• CUDA programming guide: https://docs.nvidia.com/cuda/cuda-c-programming-guide/

• CUDA toolkit documentation: https://docs.nvidia.com/cuda/index.html
• CUDA toolkit documentation - CUB: https://docs.nvidia.com/cuda/cub/index.html
• CUDA toolkit downloads: https://developer.nvidia.com/cuda-toolkit-40

• CUDA zone: https://developer.nvidia.com/cuda-zone

• Table of CUDA GPU Compute Capability: https://developer.nvidia.com/cuda-gpus

• Matching CUDA arch and CUDA gencode for various NVIDIA architectures: https://arnon.dk/matching-sm-architectures-arch-and-gencode-for-various-nvidia-cards/