Infrastructure at the Scale of Science
For more than 50 years, the Science and Technology Facilities Council (STFC) Scientific Computing centre has provided a foundation for research partnerships. It supports pioneering research communities through global collaborations with academic institutions and industrial partners. The centre combines scientific domain expertise, software engineering, and systems infrastructure to solve massive research challenges.
Modern scientific discovery is limited by the ability to manage, move, and interpret data at scale rather than by instruments alone. STFC Scientific Computing operates at the junction of world-class facilities, advanced computing, artificial intelligence, and trusted data infrastructure. Its mission centers on the technical hurdles generated by large experimental facilities like the ISIS Neutron and Muon Source and the Central Laser Facility at the Rutherford Appleton Laboratory.
Adapting to an AI-Driven Landscape
Large-scale facilities now generate continuous data streams that grow exponentially. Dr Peter Oliver, Director of STFC Scientific Computing, notes that high-performance systems must support both traditional modelling and newer AI methods. The department has developed specific AI capabilities to assist researchers in performing experiments more reliably.
The Ada Lovelace Centre represents a shift toward autonomous laboratory services. By integrating AI models into experimental workflows, researchers spend less time on manual data processing. One project, the MACE-MP model, demonstrates how AI simulates atomic behavior across solids, gases, and liquids. This allows for faster and cheaper simulations in materials science without requiring individual models for every specific chemical compound.
Managing Exabytes of Research Data
Operating across the Daresbury and Rutherford Appleton laboratories, the department manages over half an exabyte of scientific data. This infrastructure ranks among the largest in the United Kingdom. Staff maintain this storage through a matrixed approach, utilizing themes such as computational materials, cyber security, and data engineering to deploy expertise where it is needed most.
The network is built for speed. It features an internal switching capacity exceeding 30 terabits per second. This capacity connects instruments directly to storage and compute platforms. By allowing scientists to analyze data where it resides, the center avoids the logistical trap of transferring massive datasets between sites. Monthly traffic into the Rutherford Appleton Laboratory frequently reaches 600 gigabits per second.
Supporting Global Research Communities
The JASMIN platform serves as a primary example of environmental data analysis. Managed alongside the Centre for Environmental Data Analysis, it provides 90 petabytes of high-performance disk storage. Researchers use this system to monitor climate change, tracking shifts in oceans and ice sheets. It currently hosts more than 2,500 users working on over 500 distinct projects.
Looking ahead, STFC intends to leverage its national laboratory position to lead in scientific computing delivery. By merging high-performance computing, data management, and artificial intelligence, the centre aims to address societal issues from battery degradation to climate resilience. The focus remains on providing longevity for research software through the Computational Science Centre for Research Communities to ensure that findings remain accessible long after initial grants expire.