Image
Intro to GPU Programming
- Video: Introduction to NVIDIA GPU Computing
- Presentation: Present and Future of Accelerated Computing Programming Approaches
- Presentation: Inside the NVIDIA HPC SDK: the Compilers, Libraries and Tools for Accelerated Computing
- Video: ACM Winter School 2019 on HPC IIT Kanpur
Libraries
- Documentation: Math and communication libraries
- Presentation: How CUDA Math Libraries Can Help You Unleash the Power of the New NVIDIA A100 GPU
Programming Models
OpenACC
- Web Page: OpenACC.org Resources
- Forum: OpenACC Slack Channel
- Presentation: Zero to GPU Hero with OpenACC
- Presentation: Directive-Based Programming with OpenACC
- Training Series: OpenACC Training Series
- Videos: Directive Based GPU Programming
- GitHub: OpenACC Training Materials
- Docker Container: OpenACC Training Materials
- Online Course: Fundamentals of Accelerated Computing with OpenACC (Fee-based)
CUDA
- Training Series: CUDA Training Series
- Presentation: CUDA on NVIDIA Ampere GPU Architecture: Taking Your Algorithms to the Next Level of Performance
- Presentation: CUDA New Features And Beyond
- Presentation: Developing CUDA Kernels to Push Tensor Cores to the Absolute Limit on NVIDIA A100
- Online Course: Fundamentals of Accelerated Computing with CUDA C/C++ (Fee-based)
CUDA Fortran
- Documentation: Programming Guide
Python
- Presentation: CuPy Overview: NumPy Syntax Computation with Advanced CUDA Features
- Presentation: Accelerating Python with CUDA
- Video: Valentin Haenel: Create CUDA kernels from Python using Numba and CuPy
- Online Course: Fundamentals of Accelerated Computing with CUDA Python (Fee-based)
Kokkos
- GitHub: Kokkos Repository
- GitHub: Kokkos Tutorials
- Forum: Kokkos Slack Channel
- Video: Kokkos: C++ Performance Portability for Production
RAJA
- GitHub: RAJA Repository
- Video: A Tutorial Introduction to RAJA
Multi-GPU
- Presentation: Multi-GPU Programming with Message-Passing Interface
- Presentation: Multi-GPU Programming Models
- Presentation: Multi-GPU Programming with CUDA, GPUDirect, NCCL, NVSHMEM and MPI
- Presentation: A Partitioned Global Address Space Library for Large GPU Clusters
Tools
- Video: NVIDIA Nsight™ Systems "Profiling GPU Applications with NVIDIA NSight Systems"
- Video: NVIDIA Nsight Compute "Introduction to NVIDIA Nsight Compute"
- Videos: Roofline Analysis Workshop - Part 1 and 2, Part 3
- Presentation: What the Profiler is Telling You: How to Get the Most Performance out of Your Hardware
- Presentation: Optimizing CUDA Kernels in HPC Simulation and Visualization Codes Using NVIDIA Nsight Compute
- Presentation: Roofline and NVIDIA Ampere GPU Architecture Analysis
- Presentation: Performance Tuning CUDA Applications with the Roofline Model
- Video: Cross-Platform Performance Engineering with Arm Allinea Studio
- Blog: GPROF Tutorial – How to use Linux GNU GCC Profiling Tool
- Blog: Custom Application Profile Timelines with NVTX
Data Science
- Web Page: RAPIDS overview
- Web Page: RAPIDS-Getting Started (Hands-on labs and other materials)
- Forum: RAPIDS Community
- Presentations: Collection of RAPIDS Talks
- Online Course: Fundamentals of Accelerated Data Science with RAPIDS (Fee-Based)
AI/Deep Learning Libraries, Frameworks, SDKs
Introduction to AI
- Presentation: Dive into Deep Learning
- Presentation: Do-it-Yourself Automatic Speech Recognition with NVIDIA Technologies
- Online Course: Fundamentals of Deep Learning for Computer Vision (Fee-Based)
- GitHub: Deep Learning Examples (The latest deep learning example networks for training and Inference.)
CuDNN
- Documentation: CuDNN Developer Guide
- Presentation: cuDNN v8 New Advances in Deep Learning Acceleration: APIs, Optimizations, and How to Tackle the Future Challenges in Hardware and Software
- Presentation: Deep Learning Training with cuDNN
NVIDIA CLARA
- Container: NVIDIA Clara Train SDK
- Jupyter Notebooks: Intro to Clara Train SDK
- Jupyter Notebook: Clara Federated Learning
- Presentation: Clara Developer Day: Federated Learning Using Clara Train SDK
- Presentation: Clara Developer Day: Clara Train SDK Performance Walkthrough and Deep Dive
- Presentation: Clara Developer Day: Scalable and Modular Deployment Powered by Clara Deploy SDK
- Presentation: Clara Developer Day: Getting Started with Clara Train for High Performance & Iterative Experimentation with AutoML
- Web Page: NVIDIA Clara™ Parabricks Overview
DeepStream
- Presentation: Developing IVA Software Using NVIDIA DeepStream SDK
- Online Course: Getting Started with DeepStream for Video Analytics on Jetson Nano (Fee-Based)
- Online Course: AI Workflows for Intelligent Video Analytics with DeepStream (Fee-Based)
TensorRT
- Presentation: PyTorch-TensorRT: Accelerating Inference in PyTorch with TensorRT
- Presentation: Inference with Tensorflow 2 Integrated with TensorRT
- Presentation: TensorRT inference with TensorFlow 2.0
- Online Course: Optimization and Deployment of TensorFlow Models with TensorRT (Fee-Based)
TensorFlow
- Presentation: What's New In TensorFlow-An Ecosystem Update
- Presentation: Extensions of TensorFlow-Based Computational Fluid Dynamics
- Online Course: Image Classification with TensorFlow: Radiomics - 1p19q Chromosome Status Classification (Fee-Based)
Keras
- Presentation: Keras and TensorFlow: The Next Five Years
- Presentation: Porting Keras Lamba Layers to TensorFlow.js
- Online Course: Modeling Time Series Data with Recurrent Neural Networks in Keras (Fee-Based)
PyTorch
- Website: PyTorch
- Container: PyTorch
- Presentation: Latest Developments on PyTorch: Getting Started and Community Highlights
- Presentation: Dynamic Shapes First: Advanced GPU Fusion in PyTorch
- Presentation: Scaling Large Models Using PyTorch RPC
- Presentation: Named Tensors, Model Quantization, and the Latest PyTorch Features
Containers
- Video: Making Containers Easier with HPC Container Maker
- Presentation: Containers Democratize HPC
- Online Course: High-Performance Computing with Containers (Fee-Based)
Advanced
- Presentation: Inside the NVIDIA Ampere Architecture
- Presentation: Optimizing Applications for NVIDIA Ampere GPU Architecture