This week we have a number of trainings aimed at new users, training in R and an introduction to parallel programming utilising OpenMP. We also high light two training events from external sources. The ENCCS/NVIDIA workshop on Ai for Science and the EuroCC Finland workshop on porting CUDA codes to HIP which is relevant for Swedish researchers wanting to run GPU codes on the forthcoming LUMI system.

Contents of this issue

Training aimed at new users of the SNIC infrastructure:

• Online training seminar: An introduction to the UNIX/LINUX command-line, 10 February 2021 at 14:0
• Online training seminar: Cluster architecture and job submission, 17 February 2021 at 14:00

Training in R

• Online training course: "Using R in an HPC environment" in Umeå, 25-26 February 2021

Training in parallel computing

• Online training course: An introduction to shared memory parallel programming using OpenMP, 9 - 12 March 2021

Online interactive support and discussion forum:

• SNIC zoom-in - a virtual open-house: 8th February from 14:00 until 15:00

Training events from ENCCS, recommended by SNIC

• Online ENCCS/NVIDIA workshop: "AI for Science” Bootcamp, March 8-9, 2021. 09:00-12:15

Training events from EuroCC Finland, recommended by the SNIC training team

• HIP101: porting CUDA codes to HIP, 26 February 2021, 9:00 Finish time

Training aimed at new users of the SNIC infrastructure

Online training seminar: An introduction to the UNIX/LINUX command-line, 10 February at 14:00

This training seminar provides a first introduction to the UNIX/LINUX environment. It is aimed at new users of UNIX and will discuss UNIX skills required to work on a typical HPC cluster, such as provided by SNIC and LUNARC. Topics include manipulating files and directories, editing file contents and the development of simple scripts, such as required for job submission.

Time: 10 February at 14:00

For more information and registration visit the LUNARC UNIX/LINUX event page.

Online training seminar: Cluster architecture and job submission, 17 February 2020 at 14:00

This event explains key features of a contemporary HPC cluster, such as deployed at LUNARC and throughout SNIC. It will explain the principles behind the job scheduler and how the scheduler can be used to accomplish your computational work in an efficient manner. The examples will utilise the SLURM scheduler, which is deployed on the SNIC resources.

Time: 17 February at 14:00

For more information and registration visit the LUNARC cluster event page.

Training in R

Online training course: "Using R in an HPC environment" in Umeå, 25-26 February 2021

HPC2N is offering a course on R in HPC environment. The course will be two half-days, and it will be given online, through Zoom.

There is interest among R users to get started in High Performance Computing (HPC) as they need more CPU power, memory or both. Because most of the Machine Learning (ML) models in R are computationally demanding, they are a good starting point to introduce HPC concepts. In the present 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. As realistic ML models are computationally expensive when running on a Desktop, we will also teach you how to run these models on Kebnekaise, a HPC cluster at our center. The focus of the course will not be about ML though.

The course will consist of a mix of lectures and hands-ons sessions.

For more information and registration, see the HPC2N R event page. 

Training in parallel computing

Online training course: An introduction to shared memory parallel programming using OpenMP, 9 - 12 March 2021

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 within SNIC. This event is a collaboration between HPC2N and LUNARC.

This course will introduce participants to the shared-memory model for parallel programming and the OpenMP application-programming interface. In many cases OpenMP allows an existing serial program to be upgraded incrementally, starting with the parallelisation of the most time-consuming parts of the code. Typically OpenMP programs are easily ported from one shared memory multi processor system to another one.

More advanced OpenMP training events are planed during spring 2021. These will be offered by SNIC, PRACE or ENCCS and explain OpenMP tasks and GPU programming with OpenMP. This event provides an introduction to the core features of OpenMP and offers a foundation for the more specialised events.

For more information and registration see:

• HPC2N OpenMP event page
• LUNARC OpenMP event page

Online interactive support and discussion forum

SNIC zoom-in - a virtual open-house: 8th February 2021 from 14:00 until 15:00

You are invited to a virtual meeting room. Inside the meeting room we like to discuss services offered by the SNIC centres and how they can be used for your computational needs, help you process your data and visualise your results. Participants are highly encouraged to pose their own questions.

For more information and details on access, please visit the LUNARC zoom-in page.

Training events from ENCCS, recommended by SNIC

Online ENCCS/NVIDIA workshop: "AI for Science” Bootcamp, March 8-9, 2021. 09:00-12:15

This online event will take place during two half-days on March 8-9 2021 and is designed to teach researchers how AI can accelerate HPC simulations by introducing the concepts of Deep Neural Networks, including data pre-processing, and techniques on how to build, compare and improve accuracy of deep learning models. The bootcamp will introduce real-world problems, where participants learn AI concepts with step-by-step instructions using learned techniques.

For more information and registration, please visit the ENCCS event site for the bootcamp.

Training events from EuroCC Finland, recommended by the SNIC training team

HIP101: porting CUDA codes to HIP, 26 February 2021, 9:00 Finish time

With the LUMI supercomputer arriving this year, it is important for Swedish scientists who need the capabilities of LUMI to start preparing their codes. The main partition of LUMI is based on AMD GPUs, which need to be programmed in HIP instead of CUDA. The first training will be about porting CUDA applications to HIP. We confirmed with the organisers that Swedish researchers will be accepted on this course.

Please register asap, deadline is 15 February.

For more information and registration please visit the CSC event site.

Training overview

SNIC training webpage

In addition to these news letters we also provide a SNIC training webpage.

This webpage currently gives an overview on all courses currently planed. It provides links to more information and the actual registration. The webpage also includes training which is expected to be mostly of interest to individual SNIC centres. Information will be added to this page as it becomes available.