Welcome to the latest edition of the SNIC training newsletter. Today we highlight a general introduction to the systems and services provided at NSC and an application specific workshop focusing on the WIEN2k application. SNIC’s next interactive support session “zoom-in” is scheduled for the 21st October. We also highlight training events provided by the LUMI user support team and ENCCS, which are focusing on users who want to access pre-exascale systems, in particular LUMI.

• Training for new users
• Online training and support event: “Introduction to NSC”, 17 November 2021 at 10:00
• Application specific training
• Online training mini workshop: “Introduction to WIEN2k”, 18 November 2021 at 13:00
• Online interactive support and discussion forum
• SNIC zoom-in - a virtual open-house: 21st October 2021 from 14:00 until 15:00
• Training events from the LUMI support team for users preparing for LUMI access
• Online training course: OpenMP Programming, 2 - 3 November 2021
• Online training course: OpenMP offload, 9 - 10 November 2021
• Training events from ENCCS for users preparing for LUMI access
• ENCCS SYCL Workshop in collaboration with CSC and IZUM, 8 - 9 November 2021
• Training on HPDA for climate data with the Ophidia framework, November 11 @ 09:00 - 15:00 CET
• OpenMP Offloading Workshop, November 18 @ 09:00 - November 19 @ 12:30 CET
• Advanced Deep Learning with Transformers – ENCCS/RISE November 23 @ 09:00 - November 24 @ 16:00 CET
• CMake Workshop, November 29 @ 09:00 - November 30 @ 13:00 CET
• ENCCS/NVIDIA GPU Hackathon, 6th, 13th - 15th December 2021

Training for new users

Online training and support event: “Introduction to NSC”, 17 November 2021 at 10:00

An online event with two parts:

• A short introduction to NSC suitable for new users, topics include the Sigma (LiU) and Tetralith (SNIC) clusters, the module system, submitting jobs, allocating GPU:s for simulations or graphics and other useful information.
• An open session for questions with several NSC application experts attending.

Feel free to join (and leave) whenever depending on your interest. The online event is open for all present and prospective NSC users.

Time: 17 Nov 2021, 10:00 - ca. 11:00 Short introduction to NSC, ca. 11:00 - Open session

See also the NSC introduction event page, where the presentation will be available.

Application specific training

Online training mini workshop: “Introduction to WIEN2k”, 18 November 2021 at 13:00

An online training event suitable for new users of the WIEN2k software package. The first part includes a brief overview and background for using WIEN2k. Thereafter, practical examples and discussion of running WIEN2k efficiently on the NSC clusters will be made. The last part of the event will be open for hands-on testing and questions. Note that for the hands-on part you need to be covered by a WIEN2k license and have an account and compute project at NSC.

Time: 18 Nov 2021, 13:00 - ends at latest 17:00

For more information, see the NSC WIEN2k event page.

Online interactive support and discussion forum

SNIC zoom-in - a virtual open-house: 21st October 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.

Next zoom-in session: 21 October from 14:00 until 15:00

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

Training events from the LUMI support team for users preparing for LUMI access

Online training course: OpenMP Programming, 2 - 3 November 2021

OpenMP is an application programming interface designed to facilitate parallelism. It is the model of choice for shared-memory programming, which enables serial programs to be parallelised using compiler directives. Since its introduction in 1997, OpenMP has tracked the evolution of hardware used in high-performance computing including the increasing use of accelerators such as GPUs. In this course the student will learn a wide range of OpenMP topics, starting from the basics before moving to really advanced topics.

For more information and access to registration, see the LUMI OpenMP programming event page.

Online training course: OpenMP offload, 9 - 10 November 2021

This course address the use of OpenMP for programming co-processors such as GPUs. It focuses on how to get the best out of OpenMP in terms of performance by exploring the implications of possible OpenMP parallelization strategies. Advanced topics such as asynchronous execution, interoperability with CUDA/HIP and the use of multiple GPUs are covered.

For more information and access to registration, see the LUMI OpenMP offload event page.

Training events from ENCCS, recommended by SNIC

ENCCS SYCL Workshop in collaboration with CSC and IZUM, 8 - 9 November 2021

SYCL is a C++ abstraction layer for programming heterogeneous hardware with a single-source approach. SYCL is high-level, cross-platform, and extends standard ISO C++17. You will learn to:

• Use the hipSYCL compiler to generate executable for multiple hardware targets
• Write hardware-agnostic code to express parallelism using the queue, command group, and kernel abstractions
• Use buffer and accessors to handle memory across devices
• Evaluate drawbacks and advantages of unified shared memory.

For more inforation and registration, please visit the ENCCS SYCL workshop page.

Training on HPDA for climate data with the Ophidia framework, November 11 @ 09:00 - 15:00 CET

Due to the continuous increase of data volumes that many scientific domains are facing, advances in science have become dependent on the availability of scalable analysis tools able to effectively cope with these massive datasets.

This online training course aims to provide a basic understanding of HPDA challenges for eScience and how these are addressed by the Ophidia HPDA framework. Ophidia is a CMCC Foundation research effort targeting scientific data-intensive analysis, by joining HPC paradigms and Big Data approaches. The framework specifically targets the analysis on top of HPC systems and is currently involved in the ESiWACE2 CoE and the eFlows4HPC EuroHPC JU projects for large-scale scientific data analytics.

Basic knowledge of Python, Jupyter Notebooks, Linux and the NetCDF data format is required to fully take advantage of this training course. Moreover, general knowledge of climate data aspects is recommended.

Registration and more information at the ENCCS HPDA event page.

OpenMP Offloading Workshop, November 18 @ 09:00 - November 19 @ 12:30 CET

ENCCS is offering a training workshop on November 18-19 focused on GPU offloading using OpenMP. These new features, which are available since version 4.0, make OpenMP a good option for researchers and developers who wish to accelerate their code on GPUs using a cross-platform framework supported by all HPC device vendors (including NVIDIA, AMD, and Intel)

The workshop will be hands-on and interactive and lectures will be interleaved with type-along sessions and exercises. CSC Finland has contributed to the training material

Prerequisites: Previous experience with OpenMP, e.g. for conventional threading on CPUs, is not required. However, participants should be comfortable writing code in C/C++ or Fortran.

Registration and more information at the ENCCS OpenMP offloading workshop page.

Advanced Deep Learning with Transformers – ENCCS/RISE November 23 @ 09:00 - November 24 @ 16:00 CET

In recent years, Graph Neural Networks (GNNs) and Transformers have led to numerous breakthrough achievements in a variety of fields such as Natural Language Processing (NLP), chemistry, and physics. By doing away with the need for fixed-size inputs, these architectures significantly extend the scope of problems in which deep learning can be applied.

This workshop will take you from the representation of graphs and finite sets as inputs for neural networks to the implementation of full GNNs for a variety of tasks. You will learn about the central concepts used in GNNs in a hands-on setting using Jupyter Notebooks and a series of coding exercises. While the workshop will use problems from the field of chemistry as an example for applications, the skills you learn can be transferred to any domain where finite set or graph-based representations of data are appropriate. From GNNs, we will make the leap to Transformer architectures, and explain the conceptual ties between the two.

Registration and more information at the ENCCS Deep Learning event page.

CMake Workshop, November 29 @ 09:00 - November 30 @ 13:00 CET

CMake is a language-agnostic, cross-platform build tool and is nowadays the de facto standard, with large projects using it to reliably build, test, and deploy their codebases.

What you will learn

• Write a CMake build system for C, C++, and Fortran projects producing libraries and/or executables.
• Run tests for your code with CTest.
• Ensure your build system will work on different platforms.
• Detect and use external dependencies in your project.
• Safely and effectively build mixed-language projects (Python+compiled language)

Registration and more information at the ENCCS CMake workshop page.

ENCCS/NVIDIA GPU Hackathon: 6th, 13th - 15th December 2021

Together with NVIDIA and NSC, ENCCS orgnises GPU Hackathons to provide exciting opportunities for scientists to accelerate their AI research or HPC codes under the guidance of expert mentors from National Labs, Universities and Industry leaders in a collaborative environment. Teams of researchers are paired with experienced GPU mentors to learn and apply the accelerated and parallel computing skills needed by the scientific community. Both current or prospective users of large hybrid CPU/GPU HPC clusters who develop HPC and AI applications that could benefit from GPU acceleration are encouraged to participate!

The ENCCS hackathon will be designed for teams to work in productive sprint sessions with expert mentors for a full week. A wide variety of different projects are possible, for example:

• Accelerating applications written in common languages – C/C++, Fortran, Python
• Using OpenACC to gradually offload work to GPUs and incrementally improving the performance
• Identifying expensive kernels and porting them to CUDA
• Profiling already GPU accelerated applications to find hotspots for further acceleration

The event is supported by WASP and NSC through the access to BerzeLiUs system.

Please register at the ENCCS/NVIDIA GPU Hackathon page.

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 planned. 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.