This page contains news and instructional blogs for NVIDIA® Nsight™ Compute. These articles are a great resource for enhancing your understanding of all the features Nsight Compute has to offer.

GTC 2021 Lab: Optimizing CUDA Machine Learning Codes with Nsight Profiling Tools AND Analysis-Driven Optimization with Nsight Compute

In this hands-on lab, NVIDIA developers and experts will guide you through using NVIDIA's Nsight tools for analyzing and optimizing CUDA applications. You begin by using Nsight Systems to analyze the overall application structure and explore parallelization opportunities. Nsight Compute will then be used to analyze and optimize CUDA kernels, using an online machine learning code for 5G.

May 16, 2021 View the GTC Lab Session Recording | Lab Materials on GitHub

Analysis-Driven Optimization: Analyzing and Improving Performance with NVIDIA Nsight Compute

Nsight Compute is the primary NVIDIA CUDA kernel-level performance analysis tool.
In this three-part series, you discover how to use NVIDIA Nsight Compute for iterative, analysis-driven optimization.

  • Part 1 covers the background and setup needed
  • Part 2 covers beginning the iterative optimization process
  • Part 3 covers finishing the analysis and optimization process and determining whether you have reached a reasonable stopping point

January 27, 2021 Read on NVIDIA's DevBlogs

Using NVIDIA Nsight Compute 2020.1 in Containers

Containers are now ubiquitous, and for good reason; the portability and productivity enhancements they provide have made them a standard component in HPC and many other computing fields. The NVIDIA Nsight family of developer tools for analyzing performance of CUDA applications are supported in container environments. This post contains information for using Nsight Compute in container environments. Nsight Compute is specifically designed to provide detailed performance analysis of CUDA kernels running on the GPU.

August 14, 2020 Nsight Compute Overview | Read on NVIDIA's DevBlogs

GTC 2020 Lab: Modern CUDA Programming Hazards and the Linux Nsight Toolbox to Fix Them

In this hands-on lab, you'll learn from NVIDIA developers and experts about efficiently debugging, profiling, and optimizing CUDA applications on Linux. Through a set of exercises, you'll use the latest features in NVIDIA's suite of tools, including Nsight Compute 2020.1, to detect and fix common issues of correctness and performance in their applications.

May 21, 2020 Lab Instructions (PDF) | Lab Materials on GitHub

Unleashing the Power of NVIDIA Ampere Architecture with NVIDIA Nsight Developer Tools

See how NVIDIA Nsight Compute 2020.1 fits into a top-down profiling strategy and how it can help you tune your applications for use with CUDA 11.0 and NVIDIA's Ampere GPU architecture.

May 14, 2020 Nsight Compute Overview | New in 2020.1 | Read on NVIDIA's DevBlogs

Using Nsight Compute to Inspect your Kernels

NVIDIA has added Nsight Compute to the repertoire of CUDA tools available for developers. This tool is important when using newer GPU architectures. For the example project in this blog, using Nsight Compute is necessary to get the results we are after for Turing architecture GPUs and beyond. One of the main purposes of Nsight Compute is to provide access to kernel-level analysis using GPU performance metrics. If you’ve used either the NVIDIA Visual Profiler, or nvprof (the command-line profiler), you may have inspected specific metrics for your CUDA kernels. This blog focuses on how to do that using Nsight Compute.

September 16, 2019 | Nsight Compute 2019.4 (CUDA 10.1 Update 2) | Read on NVIDIA's DevBlogs

Migrating to NVIDIA Nsight Tools from NVVP and Nvprof

See how NVIDIA Nsight Compute 2020.1 fits into a top-down profiling strategy and how it can help you tune your applications for use with CUDA 11.0 and NVIDIA's Ampere GPU architecture.

July 15, 2019 | Nsight Compute 2019.3 (CUDA 10.1 Update 1) | Read on NVIDIA's DevBlogs

 Download   Documentation