NCI and NVIDIA are holding a series of bootcamps in 2022 with the focus on AI, GPU, parallel programming and speed up, big data analysis using CPU and GPU capabilities. Bootcamps and Hackathons are exciting and unique way for scientists and researchers to learn the skills needed to start quickly accelerating codes running on Gadi. We are pleased to announce three bootcamps:

NCI Australia Accelerated Data Science GPU Bootcamp

Application Deadline: March 29th, 2022

Bootcamp Dates: 1:00-4:00pm AEST, 20th and 21st April 2022

Summary

GPU Bootcamp is an exciting and unique way for scientists and researchers to learn the skills needed to start quickly accelerating codes on GPUs. Held as a virtual event across two days (with three hour sessions) , participants will learn about RAPIDS suite of open source software libraries that gives Data Scientist the freedom to execute end-to-end data science and analytics pipelines entirely on GPUs. 

In this Bootcamp, participants will be exposed to using libraries that can be easily integrated with the daily data science pipeline and accelerate computations for faster execution. This bootcamp will focus on CuDF, CuML and Dask to run analytics pipelines on multiple GPU. This online Bootcamp is a hands-on learning experience where you will be guided through step-by-step instructions with teaching assistants on hand to help throughout.

Presented by NCI Australia, the bootcamp is open to all current and prospective NCI users.

Agenda: 

Date: April 20, 2022 →1:00PM – 4:00PM Australian Eastern Standard Time

• Welcome (Moderator): 1:00 PM - 1:15 AM
• Connecting to a Cluster: 1:15 AM - 1:30 PM
• Intro to RAPIDS Ecosystem: 1:30 PM - 2:00 PM (Lecture)
• Using RAPIDS CuDF: 2:10 PM - 3:00 PM (Lecture+Lab)
• Using RAPIDS CuML: 3:00 PM - 3:30 PM (Lecture + Lab)
  
  

Date:  April 21, 2022 →1:00PM – 4:00PM Australian Eastern Standard Time

• Using RAPIDS Dask : 1:00 AM - 2:00 PM (Lecture and Lab)
• Gene Expression Classification (Mini challenge) 2:00 PM - 3:30 PM (Lecture + Lab)
• Q&A  3:30 PM - 4:00 PM 

Prerequisites

• Basic experience with Python and Data Science. No GPU programming or AI knowledge is required.​. 
• Presented by NCI Australia, the bootcamp is open to all current and prospective NCI users.
• This event has limited capacity, so please make sure that prerequisites are met before applying. 

Registration link

https://gpuhackathons.org/index.php/event/nci-australia-accelerated-data-science-gpu-bootcamp

Distributed Deep Learning GPU Bootcamp

Application Deadline: April 15th, 2022

Bootcamp Dates: 1:00-4:00pm AEST, 4th and 5th May 2022

Summary

GPU Bootcamp is an exciting and unique way for data scientists and researchers to learn the skills needed to start quickly accelerating codes on GPUs. Held as a virtual event across three days (with three hour sessions), participants will be introduced to fundamentals of Distributed deep learning and given hands-on experience on methods that can be applied to Deep learning models for faster model training. 

This Bootcamp will cover an introduction to Distributed deep learning, how to understand the System Topology and its impact on scalability and performance followed by a hands-on session with Distributed training ( Horovord, TensorFlow ). Techniques for faster convergence will also be highlighted to tackle real world problems.

Presented by NCI Australia, the bootcamp is open to all current and prospective NCI users.

Agenda: 

Date: May 4, 2022 →1:00PM – 4:00PM Australian Eastern Standard Time

• Welcome (Moderator): 1:00 PM - 1:15 AM
• Connecting to a Cluster: 1:15 AM - 1:30 PM
• Introduction to GPU Computing and Distributed Deep Learning: 1:30 PM - 2:30 PM (Lecture)
• Introduction to Profiling Tools: 2:30 PM - 3:00 PM (Lecture)
• Understanding System topology: 3:00 PM - 4:00 PM (Lecture + Lab)
  
  

Date:  May 5, 2022 →1:00PM – 4:00PM Australian Eastern Standard Time

• Distributed training with Horovod: 1:00 AM - 2:00 PM (Lecture and Lab)
• Distributed training with Tensorflow/Keras 2:00 PM - 3:00 PM (Lecture + Lab)
• Techniques for Faster Convergence 3:00 PM - 4:00 PM (Lecture + Lab)

Prerequisites

• Basic experience with Python and Deep Learning. No GPU programming knowledge is required. 
• Presented by NCI Australia, the bootcamp is open to all current and prospective NCI users.
• This event has limited capacity, so please make sure that prerequisites are met before applying.

Registration link

https://gpuhackathons.org/index.php/event/nci-australia-distributed-deep-learning-gpu-bootcamp

NCI CUDA Python GPU Bootcamp

Application Deadline: April 27th, 2022

Bootcamp Dates: 1:00-4:00pm AEST, 18th and 19 May 2022

Summary

GPU Bootcamp is an exciting and unique way for scientists and researchers to learn the skills needed to start quickly accelerating codes on GPUs. Held as a virtual event across two days (with three hour sessions), participants will learn about multiple GPU programming models using Python and can choose the one that best fits their needs to run their codes on GPUs. 

This Bootcamp will cover an introduction to GPU programming using CuPY, Numba and CUDA Python, and provides hands-on opportunities to learn how to analyze GPU-enabled applications using NVIDIA® Nsight™ Systems.

Presented by NCI Australia, the bootcamp is open to all current and prospective NCI users.

Agenda 

Date: May 18, 2022 →1:00PM – 4:00PM Australian Eastern Standard Time

• Welcome (Moderator): 1:00 PM - 1:15 AM
• Connecting to a Cluster: 1:15 AM - 1:30 PM
• Intro to GPU Computing: 1:30 PM - 2:00 PM (Lecture)
• Using Profiling Tools: 2:10 PM - 3:00 PM (Lecture)
• Accelerating on GPUs using CuPY: 3:00 PM - 3:30 PM (Lecture + Lab)
  
  

Date:  May 19, 2022 →1:00PM – 4:00PM Australian Eastern Standard Time

• Accelerating on GPUs using Numba: 1:00 PM - 2:00 PM (Lecture and Lab)
• Mini Application Challenge: 2:00 PM - 3:30 PM (Lab)
• Wrap-up and announcing winners: 3:30 PM - 4:00 PM 

Prerequisites

• Basic experience with Python. No GPU programming knowledge is required. 
• Presented by NCI Australia, the bootcamp is open to all current and prospective NCI users.
• This event has limited capacity, so please make sure that prerequisites are met before applying. 

Registration link

https://gpuhackathons.org/index.php/event/nci-cuda-python-gpu-bootcamp

Event Format

The 2-day Bootcamps will be hosted online in the Aust Eastern Standard Time (AEST) zone. All communication will be done through Zoom, Slack and email.

Contact us if you have any questions: training.nci@anu.edu.au 

A catalogue of courses to potentially offer to the community can be found here. Please submit your EOI form via this link.