events

An introduction to shared memory parallel programming using OpenMP, 9 - 12 March 2021

This online course is given in cooperation between HPC2N and LUNARC.

OpenMP provides an efficient method to write parallel programs in C, C++ and Fortran.  OpenMP programs are suitable for execution on shared memory architectures such as modern multi core systems or a single compute node of the HPC clusters deployed by Lunarc and within SNIC.

Course: Using R in an HPC environment (25-26 February 2021)

An introduction to parallel programming using Message Passing with MPI

Course: Heterogeneous computing with performance modelling

Course: Introduction to Git

Remote/online participation: The course will be given on-site, but also streamed, using Zoom. Depending on the developing situation with COVID-19, the course may become completely online. If that happens, participants will be given information about any changes to the course as well as how to connect, using Zoom.

The course will be an introduction to version control, with a focus on Git.

Course: Introduction to HPC2N, 2020-09-09

HPC2N is offering an introduction course to our systems, "Introduction to HPC2N" in Umeå, 9 September 2020.

NOTE: The course is currently being planned as both on-site and online, depending on how the COVID-19 situation develops over the summer. Regardless, there will be the option to participate remotely, through Zoom.

Scientific computing in R

This one-day, online course will help users of R get started in High Performance Computing (HPC). 

In the course, you will learn how to improve your code so that it could run more efficiently on a Desktop and also in an HPC environment. We will use ML models as one of the examples to show how to run more efficiently at an HPC cluster like Kebnekaise.

Task-based parallelism in scientific computing

ATTENTION: Due to the current situation with the covid-19 virus, the course has been postponed by at least two months. More information will follow.

The purpose of the course is to learn when a code could benefit from task-based parallelism, and how to apply it.