JuliaHLRS23

A four-day workshop that will take place in Stuttgart at the High Performance Computing Center Stuttgart (HLRS) in October 2023.

Course page: https://www.hlrs.de/training/2023/julia
Instructor: Carsten Bauer

Last years course: https://github.com/carstenbauer/JuliaHLRS22

Tentative schedule

Handout

Practical information about Hawk and the local laptops can be found in the digital handout (PDF version).

Optional: Preparing for the workshop (if you bring your own device)

Important note: We will provide preconfigured Laptops for the course and you don't need to bring your own device. If you do, please note that you won't be able to perform all exercises on your private machine, in particular those that are supposed to be done on the Hawk cluster. Also note that it won't be possible to access Hawk from your private device.

Software

What you need:

• Julia 1.9
  • I recommend to use juliaup to install and manage Julia versions!
  • curl -fsSL https://install.julialang.org | sh
• Jupyter Lab
• (Optional but recommended: Visual Studio Code)
• (Optional: LIKWID)

Workshop materials & Julia dependencies

To download the workshop materials (this GitHub repository) and to install all Julia dependencies, run

git clone https://github.com/carstenbauer/JuliaHLRS23
cd JuliaHLRS23
julia install.jl

IMPORTANT NOTE: I might make some last minute changes. To be on the safe side, make sure you update your local instance of the repository right before the start of the workshop. (You can always delete your local copy and redownload or simply git pull to update.)

Remarks:

• If you don't have Julia and want a single script to install everything you need (except Jupyter Lab and VS Code), see orga/local_install.sh.
• Because we want the workshop to be as self-contained as possible, we also install dependencies like CUDA and MPI (via Julia's Pkg). The overall size of the installation is therefore expected to be in the ballpark of 5 GB.