events

Course: Machine Learning with R in HPC

Course: Introduction to GPU programming: When and how to use GPU-acceleration?

VASP - Basic Theory and Best Practices

This one day course will start by looking at the basic theory, then go on to setting up input scripts for VASP. We will look at general recommendations for common methodologies used, some common issues when setting up a simulation, and also at running VASP on HPC2N. In addition, there will be a hands-on.

Participation is free. Coffee/tea will be provided.

Instructors: Weine Olovsson, Pedro Ojeda May, Birgitte Brydsö

Time and date: 29. October 2019, 09:00-17:00.

This whole day course will begin with a short (optional) introduction to Linux. We then continue with an introduction to HPC and HPC2N, followed by a walkthrough of the Kebnekaise system, including the GPUs and KNLs. In the afternoon we go through the module system and take a brief look at how to submit jobs. During the hands-on you will get the opportunity to try loading modules, compile a program, and submit a job to the Kebnekaise cluster. For those who are interested, there will be a guided tour of the machine rooms.

Participation is free. Coffee/tea will be provided.

Two day course focused on Tensorflow and Deep Learning. We will start with an introduction to Tensorflow, then give some basic information about Generative Adversarial Networks (GAN), and how to use GAN advanced deep learning in practice. The course will also introduce some of the recent advanced deep learning models (e.g., Transformer, ELMo, BERT) in NLP (natural language processing).

The course consists of lectures and hands-on exercises. Please bring a laptop.

Participation is free. Lunch and coffee/tea will be provided.

Python for Scientific Computing

This one-day course will give a brief, but comprehensive introduction to Python for Scientific Computing. It will have lectures interspersed with hands-on sessions where you get to try out what you have just learned.

Course: Efficient MD simulations at HPC2N

The aim of the course is to provide MD users with a set of best practices to improve the performance of their simulations. This will be accomplished by tuning the appropriate SLURM switches and flags proper to the MD packages.

Topics to be covered are:

1. Performance analysis of different versions of the packages
2. Performance analysis on different architectures (CPU, GPU, KNL)
3. General recommendations when using plugins.

Coffee/tea will be provided.

Introduction to HPC2N

We begin with a short introduction to Linux, which is what is used on HPC2N's compute clusters. This part is only meant for complete beginners to the Linux operating system, and can be omitted if you are proficient already.

The NDGF All Hands meeting was held at Umeå University this time, on Tuesday, 6 November 2018.

This is a meeting for NT1 staff, and local NDGF site sysadmins. It started out with a session for internal topics to the NT1, then after lunch there was a session for all the local site admins from around the Nordics.

More information about the meeting can be found here: https://indico.neic.no/event/19/

Local organizer for the meeting was Mattias Wadenstein, HPC2N.

The NorduGrid Collaboration invites ARC developers, system administrators and interested ARC users to our next face to face retreat in Umeå, Sweden. The purpose of the autumn retreat is to introduce the next major ARC release, the ARC 6 to site administrators and community developers.