Viking
Viking is the University’s high-performance computing facility for research projects. It’s a large cluster of interconnected computers running Linux.
Viking is hosted in a carbon-neutral data centre, EcoDataCenter. It's powered using 100% renewable energies and utilises heat recovery from hardware. From this, it transforms surplus energy into wood pellets – a clean, sustainable and renewable fuel source. This means the environmental impact of computational research has been accounted for and is reduced.
Key features
- High memory capacity handles intensive datasets.
- Large storage delivers fast processing speeds.
- Hosted in a carbon-neutral data centre powered by 100% renewable energy, utilising heat recovery and sustainable materials to minimise the environmental impact of computational research.
- Award winning: The University was recently awarded the Sustainable Digital Project or Initiative Award at the UCISA awards, with the move to EcoDataCenter forming a key part of the submission.
Spec list
- 134 compute nodes: each with 96 CPU cores per node (two processors each with 48 cores), 512GB memory, and AMD EPYC3 7643 processor generation.
- Three high memory nodes: one with 4TB and two with 2TB memory
- 60 graphics processing units (GPU): 48× NVIDIA A40 and 12× NVIDIA H100.
- Over 1.7PB storage: 1.5PB scratch space and 215TB usable NVMe.
- 100 Gb/s interconnect: with Intel omni-path architecture (OPA).
Access instructions
- You need an account to use Viking. Learn how to create a Viking account.
- If you’re not on campus, you need a virtual private network (VPN) connection.
Additional information
What it can be used for
High-performance computing powers up digital research capabilities. It can be used for a variety of purposes, including:
- scientific simulations and modelling: run complex simulations in physics, chemistry, and biology to study things like molecular dynamics and climate change.
- data analysis and big data processing: analyse large datasets from experiments or observations to find trends, patterns and correlations.
- machine learning and artificial intelligence (AI): train machine learning models and develop AI algorithms for tasks such as image recognition, language processing and making predictions.
- computational biology and bioinformatics: carry out genomic and proteomic analyses to understand biological processes and create personalised medicine strategies.
- engineering and Computational Fluid Dynamics (CFD): simulate and optimise engineering designs and study fluid dynamics in areas like aerodynamics and hydrodynamics.
Case Studies
Explore how our students and staff use high-performance computing and software in research by reading our case studies.
If you're wondering whether Viking would be useful in your project, the Research IT team can help you decide. Contact the Research IT team (part of IT Services).
Reports
Annual reports
We regularly report on this service to show how it's being used across our University community. The report includes major updates, usage statistics, research outputs and case studies and future plans.