AMD Instinct™ GPU Training

All communication will be done through Zoom, Slack and Email.

This course will give a deep dive into the AMD Instinct™ GPU architecture and its ROCm™ ecosystem, including the tools to develop or port their HPC or AI application to AMD GPUs. Participants will be introduced to the HIP (Hetero-geneous-computing Interface for Portability) programming language for AMD GPUs, other higher-level GPU languages such as OpenMP, OpenACC, and performance portable languages such as Kokkos.
In addition, there will be presentations on other important topics such as GPU-Aware MPI, and Affinity. The AMD tool suite, including the debugger, rocgdb, and the profiling tools rocprof, omnitrace, and omniperf will also be covered. A short introduction will be given into the AMD Machine Learning software stack including PyTorch and Tensorflow and how they have been used in HPC.

After this course, participants will

  • have learned about the many GPU programming languages for AMD GPUs
  • understand how to get performance scaling
  • have gained knowledge about the AMD programming tools
  • gotten an introduction to the AMD Machine learning software
  • Profiling and debugging.


Online course
Organizer: HLRS, University of Stuttgart, Germany

Start date

Sep 25, 2023

End date

Sep 28, 2023



Entry level



Code Optimization

GPU Programming

Machine Learning


Back to list

Prerequisites and content levels


Some knowledge in GPU and/or HPC programming. Participants should have an application developer's general knowledge of computer hardware, operating systems, and at least one HPC programming language.

See also the suggested prereading below (resources and public videos).

Content levels

Basic: 1 hours
Intermediate: 7 hours
Advanced: 6 hours

Learn more about course curricula and content levels.



Bob Robey, Essam Morsi, George Markomanolis (AMD)

Agenda (subject to change)

All times are CEST.
Day 1

12:45 - 13:00 Drop in to Zoom

  • 13:15 HLRS Intro
  • 13:15 AMD Presentation Roadmap
  • 13:25 Introduction to the System for Exercises
  • 13:50 Introduction to the AMD Architecture, including FPGA
  • 14:10 Overview of ROCm and Compilers, Infinity
              Hub/HPC Community
  • 14:30 Break
  • 14:40 Introduction to HIP
  • 15:40 HIP Exercises
  • 16:00 Break
  • 16:15 Porting applications to HIP
  • 16:40 Porting exercises
  • 16:55 Wrapup
Day 2 - Additional GPU Programming Languages

12:45 - 13:00 Drop in to Zoom

  • 13:00 Introduction to OpenMP®
  • 13:50 OpenMP® Exercises
  • 14:30 Break
  • 15:00 Advanced OpenMP® and Mixing HIP and OpenMP®
  • 16:00 Break
  • 16:10 Performance Portability Frameworks; Intro to Kokkos
  • 16:40 Kokkos Exercises
  • 16:55 Wrapup
Day 3 - AMD Performance Considerations and Debugging

12:45 - 13:00 Drop in to Zoom

  • 13:00 AMD Communication Fabrics and GPU-Aware MPI
  • 13:30 GPU-Aware Exercises
  • 13:45 Break
  • 13:55 AMD Node Memory Model
  • 14:35 Memory Model Exercises
  • 14:50 Break
  • 15:00 Affinity - Placement, Ordering and Binding
  • 15:40 Affinity Exercises
  • 16:00 Debuggers - rocgdb
  • 16:40 Rocgdb exercises
  • 16:55 Wrapup
Day 4 - AMD Profilers and Machine Learning

12:45 - 13:00 Drop in to Zoom

  • 13:00 Introduction to rocprof
  • 13:20 Rocprof Exercises
  • 13:30 Introduction to Omnitrace
  • 14:00 Omnitrace Exercises
  • 14:15 Introduction to Omniperf
  • 14:40 Omniperf Exercises
  • 14:50 Break
  • 15:00 Machine Learning in HPC; Introduction to ML on AMD
  • 15:30 Examples of Machine Learning projects
  • 16:55 Wrapup

Registration information

Please save the September 25-28, 2023, date for this training.

Registration will open in June, 2023, and end on September 8, 2023 at 11:59 PM.


This course is free of charge.


Khatuna Kakhiani phone 0711 685 65796, kakhiani(at)
Lorenzo Zanon phone 0711 685 63824, zanon(at)

HLRS Training Collaborations in HPC

HLRS is part of the Gauss Centre for Supercomputing (GCS), together with JSC in Jülich and LRZ in Garching near Munich. EuroCC@GCS is the German National Competence Centre (NCC) for High-Performance Computing. HLRS is also a member of the Baden-Württemberg initiative bwHPC.

Further courses

See the training overview and the Supercomputing Academy pages.