Skip to main content

This course is part of the "Practical HPC Tools" series, a series of courses presenting different HPC tools and their applications. The series in organized in collaboration with EuroCC Czechia and EuroCC Germany.

Content

JAX is a high-performance Python library for numerical computing. Although it is popular for machine learning applications, its true superpower for computational scientists lies in its features that enable efficient numerical computing. JAX provides a familiar, NumPy-style API that seamlessly compiles and runs high-performance code on GPUs and TPUs without requiring specialised parallel programmingparadigms.

This course is tailored specifically for computational scientists and HPC researchers. It introduces JAX features that benefit HPC-focused simulation codes. With practical hands-on exercises, you will explore how to improve numerical computation using JAX.

Timetable (all times in CET): 

10:00 - 10:20 Introduction to JAX Concepts and Framework 10:20 - 10:30 Exercise and Correction 10:30 - 10:45 BREAK 10:45 - 11:05 More JAX features for numerical computation 11:05 - 11:20 Exercise and Correction 11:20 - 11:35 Scaling JAX to multiple nodes 11:35 - 12:00 Q&A and Wrap up

Prerequisites

  • Basic programming skills in Python.
  • Familiarity with NumPy or SciPy.
  • Basic understanding of HPC systems.
  • Familiarity with the fundamental concepts of MPI is helpful but not mandatory.

Content Level

Beginner's content:

70%

Intermediate content:

30%

Advanced content:

0%

Community-targeted content:

0%

Language

English

Lecturers

Saheed Bolarinwa (LRZ)

Prices and Eligibility

The course is open and free of charge for people from academia and industry affiliated with an institution in the Member States of the European Union or Associated/Other Countries to the Horizon 2020 programme.

Registration

Please register with your official e-mail address to prove your affiliation.

Legal Notices

For registration for LRZ courses and workshops we use the service edoobox from Etzensperger Informatik AG (www.edoobox.com). Etzensperger Informatik AG acts as processor and we have concluded a Data Processing Agreement with them.

See Legal Notices

Withdrawal Policy

See Withdrawal

Fees

Free of charge

Pre-required logistics

Basic programming skills in Python.
Familiarity with NumPy or SciPy.
Basic understanding of HPC systems.
Familiarity with the fundamental concepts of MPI is helpful but not mandatory.