PRACE Course: Deep Learning and GPU Programming Workshop @ LRZ 2022

OVERVIEW

Learn how to accelerate your applications with OpenACC and CUDA, how to train and deploy a neural network to solve real-world problems, and how to effectively parallelize training of deep neural networks on Multi-GPUs.

The online workshop combines lectures about Accelerated Computing with OpenACC and CUDA with lectures about Fundamentals of Deep Learning for single and for Multi-GPUs.

The lectures are interleaved with many hands-on sessions using Jupyter Notebooks. The exercises will be done on a fully configured GPU-accelerated workstation in the cloud.

The workshop is co-organised by LRZ and NVIDIA Deep Learning Institute (DLI) for the Partnership for Advanced Computing in Europe (PRACE). LRZ as part of GCS is one of the currently 14 PRACE Training Centres which serve as European hubs and key drivers of advanced high-quality training for researchers working in the computational sciences.

NVIDIA DLI offers hands-on training for developers, data scientists, and researchers looking to solve challenging problems with deep learning.

All instructors are NVIDIA certified University Ambassadors.

NVIDIA Deep Learning Institute

The NVIDIA Deep Learning Institute delivers hands-on training for developers, data scientists, and engineers. The program is designed to help you get started with training, optimizing, and deploying neural networks to solve real-world problems across diverse industries such as self-driving cars, healthcare, online services, and robotics.

TRAINING SETUP

To get started, follow these steps:

1. Create an NVIDIA Developer account at http://courses.nvidia.com/join Select "Log in with my NVIDIA Account" and then '"Create Account".
2.  Make sure that WebSockets works for you:
   
   • Test your Laptop at http://websocketstest.com
   • Under ENVIRONMENT, confirm that '"WebSockets" is checked yes.
   • Under WEBSOCKETS (PORT 80]. confirm that "Data Receive", "Send", and "Echo Test" are checked yes.
3. lf there are issues with WebSockets, try updating your browser.
   We recommend Chrome, Firefox, or Safari for an optimal performance.
4. Visit http://courses.nvidia.com/dli-event and enter the event code provided by the instructor.
5. You're ready to get started. Please complete the survey at the end of the course to share your feedback.

To be able to visualise Nsight System profiler output during the course, please install Nsight System latest version on your local system before the course. The software can be downloaded from https://developer.nvidia.com/nsight-systems.

1st day Fundamentals of Accelerated Computing with OpenACC

Lecturer: Dr. Volker Weinberg (LRZ)

AGENDA (all times in CEST)

10:00-10:15  Welcome & Intro

10:15-12:00  Profiling

12:00-13:00  Lunch Break

13:00-14:20  OpenACC Directives

14:20-14:30  Coffee Break

14:30-15:45  GPU Programming and Data Management

15:45-16:00  Q&A, Final Remarks

SLIDES

• dli-intro-lrz-2022-v1.pdf
• Module1.pdf
• Module2.pdf
• Module3.pdf
• Module4.pdf
• Module5.pdf
• Module6.pdf

DOCUMENTATION

• OpenACC Webpage
• OpenACC Spec
• OpenACC 2.7 API Reference Card
• Further Resources

2nd day: Fundamentals of Accelerated Computing with CUDA C/C++

Lecturer: Dr. Momme Allalen (LRZ)

AGENDA (all times in CEST)

10:00-10:15  Introduction CUDA C/C++

10:15-12:00  Accelerating Applications with CUDA C/C++

12:00-13:00  Lunch Break

13:00-14:20  Managing Accelerated Application Memory with CUDA unified memory and nsys

14:20-14:30  Coffee Break

14:30-15:45  Asynchronous Streaming and Visual Profiling for Accelerated  Applications with CUDA C/C++

15:45-16:00  Q&A, Final Remarks

SLIDES

• Intro-CUDA
• AC_CUDA_C-whole.pdf
• NVPROF_UM_all.pdf

• NVVP-Streams-UM.pdf

• dli-wrapup-lrz-2022-v1.pdf

DOCUMENTATION

• CUDA TOOLKIT
• PRACE Best Practice Guide - GPGPU
• NVIDIA Nsight System

3rd day: Fundamentals of Deep Learning

Lecturer: PD Dr. Juan Durillo Barrionuevo (LRZ)

AGENDA (all times in CEST)

10:00-10:20  Welcome and Intro

10:20-12:00  Introduction to Deep Learning and Convolutional Neural Networks

12:00-13:00  Lunch Break

13:00-14:20  Data Augmentation, Deployment and Pre-Trained Models

14:20-14:30  Coffee Break

14:30-15:45  Advanced Architectures

15:45-16:00  Q&A

SLIDES

• FundamentalsOfDeepLearning

DOCUMENTATION

• Python

4th day: Fundamentals of Deep Learning for Multi-GPUs

Lecturer: PD Dr. Juan Durillo Barrionuevo (LRZ)

AGENDA (all times in CEST)

10:00-10:15  Intro

10:15-12:00  Stochastic Gradient Descent

12:00-13:00  Lunch Break

13:00-14:20  Introduction to Distributed Training

14:20-14:30  Coffee Break

14:30-15:45  Algorithmic Challenges of Distributed SGD

15:45-16:00  Q&A

SLIDES

• Introduction_Neural_Networks_Operations_And_Distributed_Training.pdf
• Fundamentals_of_deep_learning_for_multiple_gpus.pdf

PRACE Survey

• Please fill out the PRACE online survey under https://events.prace-ri.eu/event/1363/surveys/965 
• This helps us and PRACE to
  • increase the quality of the courses,
  • design the future training programme at LRZ, GCS and in Europe according to your needs and wishes,
  • get future funding for training events.

NEXT STEPS

Visit the NVIDIA Deep Learning lnstitute's website at https://www.nvidia.com/en-us/training/ to access more training and resources.

• Start online, self-paced training in deep learning and accelerated computing (using the account you created today).
• View upcoming workshops around the world and request an onsite workshop at your company or organization.
• Learn about the University Ambassador Program.