High Performance Computing
Science and Engineering South has a proud history of collaborating on high performance computing facilities, including Emerald.
Based at the Rutherford Appleton Laboratory in Oxfordshire, Emerald was a collaborative venture between SES universities Oxford, Southampton and UCL, along with the University of Bristol, which enabled a facility with a significantly higher specification to be developed than any of the institutions would have been able to invest in individually
At the time of its launch in 2012, Emerald was one of the most powerful supercomputers in the world. When Emerald was decommissioned in 2017, it was replaced by more powerful HPC facilities, which are listed below.
Regional Computing Centres
SES members are able to access two high performance computing centres JADE, MMM Hub and EPSRC Tier2@CSD3. If your institution is a member of the new regional centres and you would like to access that centre, please contact them directly (details below). All access options for the MMM Hub can be found via their website.
Joint Academic Data Science Endeavour (JADE)
A consortium of eight UK universities, led by the University of Oxford, and funded by EPSRC, JADE is designed for the needs of machine learning and related data science applications. There has been huge growth in machine learning in the last five years, and this is the first national facility to support this rapid development, with the university partners including the world-leading machine learning groups in at Oxford, King’s College London, QMUL, UCL, and the University of Southampton. For enquiries or to access JADE, please email JADEemail@example.com.
Materials and Molecular Modelling Hub
The MMM Hub is operated through a partnership of nine of the UK’s leading universities, with all SES members taking part in this group along with Queen’s University Belfast, the University of Kent and OCF plc. It is designed to support small to medium sized capacity computing focusing on materials and molecular modelling. For enquiries or to access the Hub, please email firstname.lastname@example.org.
Researchers at UK institutions on projects under the EPSRC remit can apply to access CSD3, which is based at the University of Cambridge, as a Tier2 HPC resource. It has also been agreed that researchers can apply for paid resources at Tier2 Centres on EPSRC research grants. Please contact email@example.com with any questions.
Below is a list of HPC systems which are available at SES institutions:
- Oxford Advanced Research Computing (ARC) is the central HPC service for the University of Oxford
- UCL Research IT Services provide access to high performance computing platforms, software and research applications, HPC training and support. UCL is now an NVIDIA GPU Research Centre.
- Southampton HPC machine Iridis is used by university researchers who are limited by the computing capability of their desktop PC, while Lyceum prepares the next generation of computational scientists to make use of HPC.
- Imperial HPC
- QMUL’s HPC cluster, Apocrita, is run by the ITS Research Group
- King’s College London’s research computing infrastructure Rosalind provides a HPC cluster and private cloud facility for compute and data intensive research.
- Cambridge’s Research Computing Services‘ HPC service is known as the Cambridge Service for Data-Driven Discovery (CSD3). It provides access to both internal researchers and external users through programmes such as EPSRC Tier2 and STFC DiRAC
HPC research communities
The High End Computing (HEC) Consortia are networks of computational researchers who have been provided with a large share of access to the national Tier-1 High Performance computing resource ARCHER2 to distribute amongst their members. Each consortium is associated to a different research area in EPSRC’s remit and also serves as a forum to share knowledge and develop software.
Some Consortia also offer access to some EPSRC Tier-2 High Performance Computing services. Details on the existing Consortia, including links to their websites and their remits can be found on the EPSRC website.