Asynchrony and CUDA Streams | CUDA C++ Class Part 2

Welcome to NVIDIA’s Modern CUDA C++ Programming Class. You will learn how to unlock the GPU’s full potential by using asynchrony and CUDA Streams. This series is for C++ developers who want to use the GPU effectively—whether you’re new to CUDA and want the fastest path from “hello world” to real acceleration, or you’re an experienced CUDA programmer ready to modernize your code with the latest best practices. If you already know C++ and want to write clean, efficient, idiomatic GPU code, this course is for you. This video is part of a broader playlist containing three videos. We advise you to start from the first video. 📝 Part 1:    • Accelerating Applications with Parallel Al...   📝 Part 3:    • Implementing New Algorithm with CUDA Kerne...   📝 Full Course:    • Modern CUDA C++ Programming Class   ➡️ Link to the slides and Google Colab to run the exercise for free on the GPU: https://github.com/NVIDIA/accelerated... For the DLI version, please visit: https://learn.nvidia.com/courses/cour... 📥 Link to download Nsight Systems locally: https://developer.nvidia.com/nsight-s... Chapters: 00:00:00 Introduction 00:00:22 Synchronous vs Asynchronous 00:08:32 Exercise Compute-IO Overlap 00:09:16 Solution Compute-IO Overlap 00:10:43 Nsight Systems 00:11:35 Exercise Nsight Systems 00:14:38 Solution Nsight Systems 00:17:01 NVTX 00:19:50 Exercise NVTX 00:20:22 Solution NVTX 00:21:19 Stream 00:35:42 Exercise Async Copy 00:36:20 Solution Async Copy 00:38:36 Pinned Memory 00:42:50 Exercise Copy Overlap 00:43:23 Solution Copy Overlap 00:44:21 Takeways