UK Research and Innovation (UKRI) announced last week that it is joining the COVID-19 High Performance Computing Consortium, which calls on government, industry, and academia to volunteer time and resources on their world-class machines for research of the virus. The collaborative body was launched in March by IBM in collaboration with the US government.
UKRI is making available several supercomputers including ARCHER, a 2.55 Petaflop supercomputer based at the University of Edinburgh.
“The UK’s Digital Research Infrastructure is playing a vital role in the global coronavirus response, from the data science of disease propagation to simulation of antibody protein structures, and the social understanding of human response to the crisis,” said Mark Walport, the UKRI's chief executive.
UK Science Minister Amanda Solloway said “By joining this consortium, our leading researchers will be able to access some of the most advanced computers in the world to speed up their research, gain access to new developments, and share the UK’s world-class computing technologies to find a solution to this virus.”
Another European institution, the Swiss National Supercomputing Centre (CSCS), is joining the consortium with Piz Daint, the sixth-ranked supercomputer in the world according to the Top 500 supercomputing list.
Piz Daint is the first large GPU-accelerated supercomputer in Europe, offering 27.15 Petaflops of peak performance. CSCS is the Consortium’s 40th member.
“We support Swiss, European and global scientific research initiatives on COVID-19. Science must be at the forefront of our common fight against this devastating pandemic,” said Professor Thomas Schulthess, director of CSCS.
There are several UK projects already using supercomputing capabilities through the consortium. One is London-based machine-learning chemistry startup PostEra, whose Moonshot Project has already identified over 20 molecular designs that target a key protein associated with COVID-19.
Another project, run by AI startup Kuano, is aimed at gaining insights from diseases similar to COVID-19 such as SARS to design a new coronavirus drug.
In addition, a team at IBM Research Europe in Daresbury is working with researchers at the University of Oxford to combine advanced molecular simulations with AI in order to discover compounds that could be repurposed as antiviral drugs.