Computer simulation and modelling are increasingly seen as the third pillar of modern science, alongside theory and experiment.
Accelerating science-based research and addressing complex problems can only be accelerated by powerful supercomputers.
Based at the University of Warwick, this brand-new high-performance computer (HPC) will create opportunities for ground-breaking research within a range of research disciplines - such as drug discovery and identifying EV battery materials. Sulis allows our partners to conduct research at enviable precision and scale.
What makes it unique?
Unlike the majority of HPC facilities, Sulis focuses on high-throughput and ensemble computing workflows to accelerate breakthroughs applicable to a range of problems in the physical sciences. These include, but are not limited to predicting material properties, yield of chemical processes, the motion of bacteria and fusion plasma stability.
Sulis uses CPU and GPU hardware, configured to support research applications that rely on large ensembles of small scale calculations to enable greater quantities of data sampling, and hence accurate estimation of the uncertainty in predictive models. By running thousands of simulations concurrently in high-throughput mode, time is dramatically reduced, increasing the scope and ambition of research of research that can be accomplished in a given time.
This is the first machine of its kind in the UK to focus on enhancing ensemble computing workflows, addressing an unmet need in the academic research sector.
Servers housing some of the 90 Nvidia A100 GPUs in the Sulis system
How does the computer support the University's sustainability ambitions?
Sulis is housed in a purpose-built energy-efficient data centre at the University of Warwick.
An upgrade to the existing power and cooling infrastructure was undertaken to support additional load, increase the site’s efficiency and maintain redundancy.
A dedicated HPC space was created by expanding into a space reserved for this purpose in the building's design, including replacement of the entire UPS system for a more efficient solution.
Power Usage Effectiveness
Power usage effectiveness (PUE) is a ratio that describes how efficiently a computer data centre uses energy - specifically, the amount of energy that is used by the computing equipment itself (in contrast to cooling and other overhead that supports the equipment).
PUE is the ratio of the total amount of energy used by a computer data centre to the energy delivered to computing equipment. The closer to 1 the PUE is, the better in terms of energy efficiency. Sulis operates at a PUE of 1.12, well ahead of typical cloud providers at who typically operate at a PUE around 1.25. Efficiency is even higher in the winter months when circulated water is cooled purely by exposure of the heat exchanger to the outside air temperature.
Sulis is administered and supported by experienced HPC specialists based within the Scientific Computing Research Technology Platform at the University of Warwick
Who will benefit?
Chemists, material physicists and astronomers are some of the researchers who will benefit from this brand-new facility.
More higher education researchers than ever before will have access to state-of-the-art ensemble computing/parallel computing equipment based at the University of Warwick.
Sulis creates training and development opportunities for early career researchers and academics who want to increase the scale of their research with access to hitherto unavailable capacity for high-throughput computing.
It will also support Midlands-based PhD students in their research, with approximately 500 users expected to benefit from the facility during its lifetime
How was it developed?
Funded by the UKRI World Class Laboratories fund, and the HPC Midlands+ Consortium, Sulis has been piloted with Midlands Innovation researchers from Aston, Birmingham, Loughborough and Nottingham universities.
The HPC data centre was constructed by specialists 2bm. The servers, networking and storage equipment which constitutes the Sulis supercomputer were supplied by OCF Limited . The system is operated by the Scientific Computing Research Technology Platform at the University of Warwick.
Sulis is now the largest facility of its kind in the Midlands, replacing the previous HPC Midlands+ system Athena which was hosted at Loughborough.
It complements other UK Tier-2 HPC systems which primarily focus on providing a resource for researchers to run smaller numbers of very large calculations.
Professor David Quigley, Director of the Scientific Computing Research Technology Platform and academic lead on the Sulis project.