GPU Profiling (Performance Timelines: Rocprof and Omnitrace): Part 4 of HIP Training Series
AMD will present a multi-part HIP training series intended to help new and existing GPU programmers understand the main concepts of the HIP programming model. HIP® is a parallel computing platform and programming model that extends C++ to allow developers to program GPUs with a familiar programming language and simple APIs. Each part will include a 1-hour presentation and example exercises. The exercises are meant to reinforce the material from the presentation and can be completed during a 1-hour hands-on session following each lecture.
This training series is open to OLCF and NERSC users via Zoom. OLCF users will be using HIP for AMD GPUs on Frontier. NERSC users will be using HIP for Nvidia GPUs on Perlmutter. Please note that participants will register for each part of the series individually. The exercises for the hands-on portion can be found in this GitHub repository. The Q&A for all the sessions will be in this Google doc.
Part 4: GPU Profiling (Performance Timelines: Rocprof and Omnitrace)
10 a.m. - 12 p.m. (Pacific Daylight Time/UTC -7), Monday, October 2, 2023
Application developers have found that viewing performance data as a timeline of events and communications can yield important insights. We’ll look at how to use the AMD rocprof and Omnitrace tools to present performance timelines and the kind of data that can be collected and shown. Rocprof enables the collection of hardware counter data to display on the timeline. Omnitrace adds to the GPU hardware data many other data sources on the GPU and the CPU. Timeline analysis will help to understand the performance of streams on the GPU. And it becomes critical to understanding and tune the tight cooperation between GPU and CPU in many applications. Hands-on exercises will run the analysis tools and generate timelines to analyze.
Please register online for the remote-only event.