The next HPC Champions meeting will be held virtually on the afternoon of the 15th December 2020, and will involve talks and discussion sessions. We will confirm topics closer to the time but are likely to include updates from supercomputing services, training, and HPC software. The workshop will run between 1pm and 5pm GMT. If you have an idea for a discussion topic then please get in touch with the organiser: Andy Turner (a.turner@epcc.ed.ac.uk).
HPC Champions is a network of individuals with the goal of supporting HPC users in the UK to learn about HPC, find the most appropriate system for their purposes, and to get the most out of that system. The roughly bi-annual HPC Champions workshops play a key part in this, increasing the coordination between the different tiers of HPC in the UK and acting as a launching point for collaboration on projects related to training, outreach, reference materials and more.
The network includes a wide number of HPC communities including RSEs and system administrators from the local, regional and national supercomputing centres, as well representatives of HPC consortia, and we encourage anyone with an interest in supporting researchers and RSEs to make the best possible use of the HPC resources to attend. HPC Champions has been co-located with both the RSE Conference and HPC-SIG in the past, and this time is partnered with the ARCHER2 UK national supercomputing service.
If you have any questions, please contact the organiser: Andy Turner (a.turner@epcc.ed.ac.uk).
Participants are required to abide by the ARCHER2 Training Code of Conduct.
Draft schedule:
- 1300-1440: Presentations and lightning talks (see below for details)
- 1440-1510: Break
- 1510-1530: Selection of discussion topics
- 1530-1630: Discussion session
- 1630-1645: Wrap-up and planning for next meeting
Presentations
1300-1315: Welcome
Andy Turner, EPCC, The University of Edinburgh
1315-1400: European Environment for Scientific Software Installations (EESSI)
Kenneth Hoste, University of Ghent
What if there was a way to avoid having to install a broad range of scientific software from scratch on every workstation, HPC cluster, or cloud instance you use or maintain, without compromising on performance?
The European Environment for Scientific Software Installations (EESSI, pronounced as “easy”) is a brand new collaboration between different European HPC sites & industry partners, with the common goal to set up a shared repository of scientific software installations that can be used on a variety of systems, regardless of the operating system or processor architecture of the client system, or whether it’s a full size HPC cluster, a cloud environment, or a personal workstation.
- website: https://www.eessi-hpc.org
- GitHub: https://github.com/EESSI
- documentation: https://eessi.github.io/docs
1400-1420: Isambard 2: Adding SVE to the World’s First Production ARMv8 Supercomputer
Andrei Poenaru, University of Bristol
The Isambard Tier-2 HPC service is the world’s first supercomputer to use ARMv8 in production. In September 2020, the existing Arm and multi-architecture (MACS) partitions were doubled in size, and Isambard 2, a 72-node Fujitsu A64FX partition, was added. This talk presents the University of Bristol HPC Group’s experience with setting up and running early workloads on the SVE-enabled A64FX chips in Isambard 2.
1420-1430: Baskerville: A National Tier 2 Accelerated Compute Facility
Simon Branford, University of Birmingham
This collaboration between the University of Birmingham, The Rosalind Franklin Institute, The Alan Turing Institute, and Diamond Light Source, will provide a state-of-the-art platform for GPU-accelerated computing. This talk will introduce the facility and resources it will provide.
1430-1440: Sulis: An EPSRC platform for ensemble computing delivered by HPC Midlands+
Matt Ismail, University of Warwick
This talk will briefly outline “Sulis”, a recently-funded Tier-2 HPC service aimed at non-traditional parallelism in the form of ensemble computing.