Warm welcome to the first edition of the new NAISS training newsletters.

In this edition we highlight the upcoming CodeRefinery workshop, training events organised by individual Swedish HPC centres and training events organised by ENCCS.   These events should be of interest to members of the NAISS community.

CodeRefinery workshop is highly recommended to community members writing and maintaining their own codes.  For new users of HPC the course on R, Python and Julia and the UPPMAX introduction course should be of interest.  Users with the need to run containers should check out the singularity workshop.

The first two ENCCS events are more advanced, addressing users with an interest in GPU computing.   The creative AI webinar, will be on a general interest level suitable for a general audience.

Overview

CodeRefinery training

• CodeRefinery workshop, March 21-23 and 28-30, 2023

Training events organised by HPC centres in Sweden

• Online introduction to running R, Python, and Julia in HPC, 8-10 February 2023
• UPPMAX Introductory Course, February 13-16, 2023 (online)
• Online training workshop: Basic Singularity: “Running and building Singularity containers”, 10 March

ENCCS training events

• SciML GPU Bootcamp, 21-22 February 2023, 09:00 - 13:00 CET
• AMD GPU Hackathon, 7 and 14-16 March, 2023
• Creative AI Webinar, February 24, 09:00 – 12:00 CET

CodeRefinery training

CodeRefinery workshop, March 21-23 and 28-30, 2023

Do you write code for your research? Do you manage data? Most importantly, are you doing it properly? Has it ever happened to you to have to rewrite or simply copy-paste codes from one folder to another, possibly losing things during the process? Have you ever opened an old code you wrote and had no clue what it does? Maybe you feel you waste a lot of time doing a lot of manual work to process your data and would like to automate it? Or maybe you also are not doing any of this yet, but would like to avoid these issues from happening in the future and do things properly from the beginning?

If you answered yes to any of the previous questions, then you are strongly encouraged to participate in the CodeRefinery workshop on March 21-23 and 28-30, 2023.

Topics covered in the workshop include:

• Using Git to maintain your codes and collaborate with others
• Principles of reproducible research and FAIR
• Good coding practices in general (writing documentation, testing, how to choose an appropriate license)
• The Workshop will not teach you to code, but all the "good practice skills" that are needed to write good maintainable software and achieve reproducible research.

The workshop will alternate type along tutorials given by instructors to break-out room exercises where you will get to practice. Each breakout room will have an exercise leader to help you. You can also register as a team, in which case the team will be in the same break out room (you can also exercise together in-person if you prefer), each team can propose their own exercise leader otherwise they will be assigned one. If you already have some experience and would like to strengthen your skills and help others, you can also volunteer as an exercise leader. Some places also offer in-person exercise sessions, check if there is a session offered near you from the workshop page below.

More information and the registration form may be found on the CodeRefinery workshop page (GitHub).

Training events organised by HPC centres in Sweden

Online introduction to running R, Python, and Julia in HPC, 8-10 February 2023

Learn how to run R, Python, and Julia at Swedish HPC centres. We will show you how to find and load the needed modules, how to write a batch script, as well as how to install and use your own packages, and more.

The course will consist of lectures interspersed with hands-on sessions where you get to try out what you have just learned.

This course will consist of three half-days, one for each language. Participation is open to all current and prospective users of SNIC/NAISS.

Time and date: 2023-02-(8-10)

Location: Online, Zoom

For more information and registration, see the R, Python, and Julia course website (HPC2N).

UPPMAX Introductory Course, February 13-16, 2023 (online)

This 4-day course provides an introduction to the high-performance computing (HPC) environment offered at UPPMAX.  It offers an overview on how the systems work and covers skills from basic Linux usage to more advanced tips and tricks, including Python scripting.

Participation is open to all existing and prospective UPPMAX users.

When: February 13-16, 2023.

Where: online via Zoom.

For more information and registration, please visit the UPPMAX introductory course page.

Online training workshop: Basic Singularity: “Running and building Singularity containers”, 10 March

When: 9:15-12:00, 13:15-16:00 - 10 March 2023.

Where: online via Zoom.

The online workshop is an introduction to the basic concepts of containerised software environment solutions within the Singularity framework (Sylabs).

During the workshop you will have the opportunity to follow the interactive guide on

• how to run Singularity containers
• how to build your own
• good (and bad) practices on designing and building Singularity recipes
• build and/or host container remotely and what are the limitations

To fully benefit from the workshop, basic Linux system administration experience is highly desirable i.e. knowledge of package management and common tools for building and managing software: git, pip, conda, wget, curl …

Please follow the instructions to install Singularity (UPPMAX) on your computer before the workshop.

For more information and access to registration, please visit the singularity workshop page (UPPMAX).

ENCCS training events

SciML GPU Bootcamp, 21-22 February 2023, 09:00 - 13:00 CET

Scientific machine learning (SciML) is an emerging multidisciplinary field which includes elements from applied and computational mathematics, computer science and the physical sciences. In particular, physics-informed neural networks (PINNs), a class of Deep Learning (DL) networks, can be applied to scientific applications solving linear and non-linear equations with demanding accuracy and computational performance requirements. PINNs are specifically designed to integrate scientific computing equations, such as Ordinary Differential Equations (ODE), Partial Differential Equations (PDE), non-linear and integral-differential equations into the DL network training.

This online event will take place during two half-days on 21-22 February 2023 and is designed to teach researchers and engineers how SciML can address domain-specific data challenges and extract insights from scientific datasets. The Bootcamp is co-organised by ENCCS, OpenACC.org and NVIDIA.

Further information and registration at the SciML GPU bootcamp page (ENCCS).

AMD GPU Hackathon, 7 and 14-16 March, 2023

Dardel is a Swedish flagship supercomputer operated by PDC at KTH in Stockholm. It is an HPE Cray EX system with AMD EPYC™ 64 core CPUs and AMD Instinct MI250X GPUs. Compute time on Dardel is available for academic research via SNIC.

This hackathon, which will be held in-person at KTH in Stockholm, is for teams of research software developers who want to accomplish one or more of the following goals:

1. Port their code to run on AMD Instinct™ GPUs using any of the GPU programming frameworks.
2. Improve the performance of code that already runs on AMD Instinct™ GPUs
3. Scale up their codes to multiple AMD Instinct™ GPUs

During the hackathon, project teams will be paired up with expert mentors from HPE, AMD, PDC and ENCCS who will guide them during the hackathon to accomplish their stated goals. Teams attending the event will be given access to the Dardel system ahead of the hackathon. Each team should have at least two members who are prepared to work full-time on their projects for the duration of the hackathon. We furthermore also strongly encourage diversity in project teams.

Further information and project submission at the AMD GPU hackathon page (ENCCS).

Creative AI Webinar, February 24, 09:00 – 12:00 CET

The last year has seen an explosion in “creative Al”, in particular, the performance of text generation models (such as chatGPT) and image generation models (such as Stable Diffusion, DALL-E-2 and Midjourney) has left many outside of the field shocked by the suddenly impressive capabilities of Al.

In this three-hour webinar, we will give an overview of where these models come from, how they work and what can be expected in the future.

Registration and further information at the creative AI webinar page (ENCCS).