High performance computing

High performance computing (HPC) allows our researchers to find results that would take years of computations on standard computers within hours or days.

Our magnetic resonance imaging (MRI) and magnetoencephalography (MEG) scanners have demanding computing requirements, in terms of data storage, data backup and processing capabilities.

Typically, data analysis times are at least 10 times greater than the acquisition time. So provision of an appropriate and scalable IT environment is essential for delivering optimal research productivity, testing new ideas and pushing forward new hypotheses on the functioning of the brain.

In addition, the precious data we acquire are curated and shared to facilitate national and international collaboration within our research area.

Largest cluster dedicated to neuroimaging

This is all made possible by a high-performance compute cluster coupled to a large storage array and fast quad-core workstations.

Our cluster is the largest in the UK that is completely dedicated to neuroimaging applications. It consists of more than 100 compute nodes, each equipped with 12 CPU cores and 192GB of RAM. A conservative estimate puts the performance of our cluster at 14 TFLOPS (trillion operations per second).

Collaborative, coherent working environment

Our services include a remote desktop interface that allows researchers and collaborators to connect to our facility from any internet-connected computer. This allows us to keep all data within the Centre and provide a coherent working environment.

The service is hosted on a dedicated pool of virtual machines. Together with the cluster and the workstations, we also house a 1PB (1,000TB) high-capacity, multi-site redundant storage system.