Höchstleistungsrechenzentrum Stuttgart: Machine Learning with AMD GPUs and ROCm Software

Prerequisites and content levels

Prerequisites:

Familiarity with Linux operating systems, including Linux shell (training will use Ubuntu)
Access to an SSH client to enable remote access for interactive portions of the training
Working proficiency in English (all training will be conducted in English)
Basic understanding of machine learning/deep learning concepts

Familiarity with TensorFlow and Pytorch will be a plus. Tutorial to explore prior to the course: "Learn and use ML" section: www.tensorflow.org/tutorials, a "Deep Learning with PyTorch" section: pytorch.org/tutorials, and a Python tutorial. In suggested prereading (videos and resources below) you will find more AMD material.

Please make sure to have a functioning working environment and remote access for interactive portions of the training prior to the course. In case of questions, please contact the course organizer (see below).

Content levels:

Intermediate: 1 hour
Advanced: 3 hours

Learn more about course curricula and content levels.

YouTube videos

Resources

Instructors

Derek Bouius (AMD), Will Wang (AMD), and Pak Lui (AMD)

Agenda

This course will be provided using Microsoft Teams. All times are CET.

13:30 - 14:00 Login and Microsoft Teams setup

14:00 - 14:15 Welcome - HLRS, AMD

14:15 - 14:45 Instinct and ROCm product overview

AMD Instinct GPUs
ROCm software overview
Software ecosystem supporting Instinct and ROCm
- ML frameworks

14:45 - 15:30 ROCm basics

ROCm installation
Basic ROCm tools

15:30 - 15:35 Pause

15:35 - 16:35 Machine learning with AMD GPUs

Introduction to DL/ML on ROCm
TensorFlow installation and testing
PyTorch installation and testing
ML with multiple GPUs
ResNet50

16:35 - 16:45 Pause

16:45 - 17:45 Additional ROCm topics

Docker
HPC Job Scheduler and Monitoring
Math Libraries
Communication Libraries for Multi-GPU Scale-out

17:45 - 17:50 Pause

17:50 - 18:20 GPU performance profiling

18:20 - 18:30 Closing

Registration information

Registration is closed.

If the course is full, then please register to the waiting list, so that we can inform you when we can provide a second run of this course.

Fees

This course is free of charge.

PRACE PATC and bwHPC-C5

HLRS is part of the Gauss Centre for Supercomputing (GCS), which is one of the six PRACE Advanced Training Centres (PATCs) that started in Feb. 2012.
HLRS is also member of the Baden-Württemberg initiative bwHPC-C5.
This course is not part of the PATC curriculum and is not sponsored by the PATC program.

EXCELLERAT

This workshop is part of the collaboration between AMD and the Horizon-2020 Centre of Excellence EXCELLERAT. AMD is an EXCELLERAT Interest Group. See also the EXCELLERAT Service Portal for more information.

Contact

Rolf Rabenseifner phone 0711 685 65530, rabenseifner(at)hlrs.de
Khatuna Kakhiani phone 0711 685 65796, kakhiani(at)hlrs.de
Lorenzo Zanon phone 0711 685 63824, zanon(at)hlrs.de

Official course URL

http://www.hlrs.de/training/2021/DL1-AMD

Further courses

https://www.hlrs.de/training/ and https://www.hlrs.de/training/overview/