ARCHER2 Weekly Newsletter


Spring 2025 UKRI Access to HPC Call open

Deadline: 22 May 2025 4:00pm UK time

ARCHER2 access page :

Full details of UKRI call :

Webinar for potential applicants

UKRI will hold a webinar on 29 April 2025 from 11:00am to 12:00pm UK time. This will provide more information about the funding opportunity and a chance to ask questions.

Register for the webinar (Teams) :

UKRI National Federated Compute Services Network+ Flexible Funding Call

Deadline: Noon on Friday 16th May 2025

We are pleased to announce the opening of the Flexible Funding Call from the National Federated Compute Services NetworkPlus (NFCS) team.

We invite applications for the flexible funding call. The Flexible Funding call is designed to support projects that: (a) collect relevant evidence and documentation to contribute to the development of the federation roadmaps; (b) contribute to building the community around NFCS.

The flexible nature of the call will enable the selection of projects on various scales that span the three pillars of the NetworkPlus: Community, Technology, and Governance. The selection of projects will seek to achieve a balance across these pillars and between large and small projects. The Flexible Funding call is the principal mechanism through which the community can contribute to developing the insights needed for the roadmaps, and the project selection will look to ensure that a range of communities of different sizes and maturities with respect to large-scale computing and data services are engaged.

The call documents (application form, call guidance and draft collaboration agreement) can be found at:
https://www.archer2.ac.uk/community/nfcs/

To ensure prompt delivery of funds, please spend time reviewing the Draft collaboration agreement and share with the relevant team at your institution: Draft Collaboration Agreement

Deadline for applications for the call is noon on Friday 16 May 2025.

If you would like to be kept informed of the work of the NFCS Network+, please register for the National Federated Compute Services Network+ mailing list

We are looking to recruit Technical Coordinator for the project. The job description can be found at: https://www.jobs.ac.uk/job/DMH951/research-programme-technical-coordinator and the deadline is 13th April 2025.

Finally, the slides (and soon videos) from the launch event for the NFCS are now available at: UKRI Federated National Compute Services NetworkPlus London

Exhibits and Engagement Activities for the ARCHER2 Celebration of Science 2025

We are looking for exhibits and/or activities to participate in the ARCHER2 Celebration of Science 2025 on the 14th - 15th May 2025 in Edinburgh at South Hall, The University of Edinburgh. These exhibits/activities will be in the main foyer area of the event and attendees will be able to engage with these throughout the event. These exhibits may for example showcase your outreach activities aimed at the general public or showcase dissemination activities for fellow professionals. These do not need to be on a large scale, our aim is to highlight a diverse range of activities.

If you are interested in showcasing your activities at the event please email events@archer2.ac.uk to discuss.

ARCHER2 Celebration of Science 2025

Following the success of ARCHER2 Celebration of Science 2024, we are pleased to announce that ARCHER2 Celebration of Science 2025 will take place on the 14th - 15th May 2025 in Edinburgh at South Hall, The University of Edinburgh. The event will provide ARCHER2 users with the opportunity to showcase their science achievements on ARCHER2. The two-day programme will include a variety of talks from ARCHER2 researchers and other invited presenters involved in the ARCHER2 service, a poster session and a panel session.

The event is in-person and there is no registration fee. Registration deadline is 29th April 2025.

Full details and registration forms

GPU programming using directives

Online, 22, 24, 28 and 30 April 2025 10:00 - 12:00 BST

The OpenMP offload and OpenACC APIs support portable GPU programing using directives in Fortran, C and C++. This course will provide an introduction to GPU programming with directives, aimed at scientific application programmers wishing to develop their own software. The course will give a background on GPU architecture and programming API, and then focus on using OpenMP offload directives. The course will cover offloading code and data movement, as well as more advanced topics such as interoperability with HIP and performance tuning. We will also cover the similarities and differences between OpenMP and OpenACC.

Note that although the course will use AMD GPUs for practical exercises, it is also appropriate for those wishing to use NVIDIA or Intel GPUs.

Full details and registration

Modern C++ for Computational Scientists

Durham, 23 - 24 April 2025 09:30 - 16:00

With the recent revisions to the C++ language and standard library, the ways it is now being used are quite different. Used well, these features enable the programmer to write elegant, reusable and portable code that runs efficiently on a variety of architectures.

However it is still a very large and complex tool. This course will cover a minimal set of features to allow an experienced non-C++ programmer to get to grips with language.

These include:

  • overloading
  • templates
  • containers
  • iterators
  • lambdas
  • standard algorithms

We will also briefly cover some important libraries for numerical computing.

Full details and registration

Green software use on HPC

Edinburgh, 13 May 2025 10:00 - 15:30 BST

This lesson introduces environmental sustainability principles in the context of high performance computing (HPC) systems. Understanding the scale of emissions from different sources is critical to being able to make changes to work in a more environmentally sustainable way. This lesson will give you the ability to understand emissions arising from use of HPC system and how to quantify them. We will use practical examples and real data from an existing HPC facility to illustrate the concepts. By the end of this lesson you should:

  • Be able to understand the concept of carbon efficiency and how it relates to energy efficiency including cases where energy efficiency can be at odds with carbon efficiency
  • Know about how HPC systems potentially reduce emissions
  • Understand carbon intensity of electricity generation and the implications for carbon aware use of HPC
  • Appreciate the embodied emissions associated with HPC hardware and how they impact carbon aware use of HPC
  • Be aware of the frameworks used to measure and report on carbon emissions and how the terms used in these frameworks map onto HPC
  • Gain practical advice on how you can measure and improve the carbon efficiency of your use of HPC

We use the UK National Supercomputing Service, ARCHER2 as an example throughout this lesson but the principles and learning should be applicable to any HPC system.

Full details and registration

AMD MI300 Series Hackathon : Registration deadline 7th April 2025

Edinburgh, 27 - 29 May 2025 10:00 - 16:00 BST

ARCHER2 and AMD are hosting a three-day in-person hackathon to explore the potential of AMD GPUs for research. The event will focus on porting and optimizing code using the latest AMD development suite. This event is ideal for both expert practitioners and less experienced users who want to convert CPU-focused codes into efficient GPU-ready applications, and enhance the performance of current GPU-based codes using AMD’s next-gen MI300x GPU cards or the new MI300A APU superchip.

We’re looking for teams of 2-5 members. Participants will work with AMD experts and have access to AMD resources. In-house experts from the ARCHER2 team are also on hand.

Deadline for applications is extended to 7th April 2025. Full details and registration Limited spaces: Applications will be reviewed as received, so early submission increases your chances.

Recently added known issues

The “Known Issues” page of the ARCHER2 Documentation https://docs.archer2.ac.uk/known-issues/ lists all current open known issues including a description of the issue, its symptoms and any work-arounds.

No recent issues.

Upcoming ARCHER2 Training

Further details of upcoming training

We always welcome researchers wishing to present their work in a webinar - please contact the Service Desk if you would be interested in presenting your work.

Twitter

Recordings of past courses

Recordings of past virtual tutorials