SpinENGINE

SpinENGINE is an EU-funded programme that brings together leading research groups from across Europe to explore massively-parallel computing frameworks based on the emergent behaviour of nanomagnetic ensembles. Project partners include researchers from Norwegian University of Science and Technology (NTNU), Ghent University, IBM Zurich and ETH Zurich.

Magnetic systems such as artificial spin ices and nanoring arrays show highly-complex, non-linear and tunable magnetic properties that can be exploited for realising physical reservoir computing (RC) , which is a biologically-inspired computational approach and shows promise for processing of temporal or sequential data. 

The FMM group at the University of Sheffield are concentrating on the nanoring implementation of physical RC. An example device is shown in the SEM micrograph below. The group has fabricated numerous devices for neutron, X-ray, magneto-optical and magnetoresistive measurements. The interconnected rings show a rich combination of magnetic states when a rotating magnetic field is applied, as shown in an image obtained using X-ray Photoemission Electron Microscopy (X-PEEM) below. This creates a strongly non-linear response of the whole ring array, which can be exploited to perform RC.

SpinENGINE aims to fabricate a proof-of-concept device capable of solving pattern recognition and classification problems, and produce a roadmap for the further development of magnetic RC. Success in the SpinENGINE project will have vast implications for data analysis at all scales, ranging from low power computation in the simplest sensor node to accelerated data processing in the most complex supercomputer.

The related EPSRC-funded project MARCH aims to explore the parameter space of the nanoring system, making it possible to generate novel configurations suited to a range of tasks, such as recognising spoken digits. We also collaborate with the partners of MARCH for more efficient realisation of reservoir computing.

More details of the SpinENGINE project can be found at www.spinengine.eu.

SpinENGINE is funded by the European Union’s Horizon 2020 FET-Open programme under grant agreement No 861618.