Blue Waters User Portal |

Introduction to NVIDIA Nsight Compute - A CUDA Kernel Profiler

Presenter

Magnus Strengert

Senior Software Engineer
NVIDIA

Bio

Magnus Strengert is a Senior Software Engineer at NVIDIA. He is the technical lead for the compute kernel profiling tools. His current focus is on NVIDIA Nsight Compute, a CUDA kernel profiler that supports developers in analyzing and optimizing GPU kernel code. Magnus obtained his PhD degree in Computer Science from the University of Stuttgart in 2010. He joined NVIDIA in 2008.

Abstract

Understanding and optimizing the runtime behavior of your code can be a challenging effort but is often rewarded with significant performance gains. NVIDIA Nsight Compute is a CUDA kernel profiler that provides detailed performance data and offers guidance for optimizing your CUDA kernels. You'll learn about how to collect a wide range of performance data for your CUDA kernels, how automatic rules help in detecting common performance pitfalls and offering guidance through the profile reports, how to quickly compare profiling results to evaluate the effects of your code changes, and how to customize the tool to fit best to your optimization workflow.

Session details

When: 11:00 CST, November 6, 2018
Length of session: 1 hour
Target audience: The webinar is intended for anyone actively writing or being interested in writing CUDA kernels.
Prerequisites: This webinar assumes basic knowledge of CUDA and GPUs.
Software Availability: NVIDIA Nsight Compute is installed as part of the NVIDIA CUDA Toolkit, starting with version 10.0. The tool can also be downloaded separately at https://developer.nvidia.com/nsight-compute after registering with the NVIDIA Developer program.
Software Requirements: Please see the sections “Platform Support” and “GPU Support” in the product documentation at https://docs.nvidia.com/nsight-compute/ReleaseNotes/index.html#support for the latest information.

Webinar video

URL: https://youtu.be/nYSdsJE2zMs