[Online] Performance Analysis on GPUs with NVIDIA Tools

Wednesday Oct 9, 2024, 9:00 AM → 1:00 PM Europe/Berlin

online

Date and Time

The course will be held online on October 9th from 9 am to 1 pm.

 

Prerequisites

• Basic understanding of GPU programming with CUDA, OpenMP, or OpenACC
• Compiling applications by invoking the compiler on the command line
• Logging in to a remote machine via SSH and handle a Linux command line remotely

 

Learning Objectives

Porting Code to the GPU promises large speedups, but there can be pitfalls on the way to realizing its potential. This course will introduce NVIDIA's profiler as a tool to spot common performance bugs that arise when porting code to the GPU.

A practical demonstration introduces the basics of GPU performance analysis. NVIDIA's profiling tool Nsight Systems ist used to analyze GPU utilization and to spot performance anomalies. The complementary tool Nsight Compute is then used to gain more insight about the performance of individual GPU kernels. The performance analysis will be guided by simple, resource-based performance models which will enable the developer to develop a concept of how far the performance is from the "target."

Attendees will be able to follow along the demos and conduct their own experiments on the NHR@FAU GPU cluster.

 

Certification

All actual course attendees will receive a course certificate.

 

Language

The course will be held in English.

 

Instructors

Dominik Ernst, Dr. Sebastian Kuckuk.

The course is organized by the Erlangen National High Performance Computing Center (NHR@FAU).

Prices and Eligibility

The course is open and free of charge for participants from academia from European Union (EU) member states and countries associated under Horizon 2020.

 

Withdrawal Policy

Please only register for the course if you are really going to attend. No-shows will be blacklisted and excluded from future events. If you want to withdraw your registration, please send an e-mail to sebastian.kuckuk@fau.de.

 

Wait List

To be added to the wait list after the course has reached its maximum number of registrations send an e-mail to sebastian.kuckuk@fau.de with your name and university affiliation.