The Fun-COMP project aims to develop a new wave of industry-relevant technologies that will extend the limits facing mainstream processing and storage approaches. We will do this by delivering innovative nanoelectronic and nanophotonic devices and systems that fuse together the core information processing tasks of computing and memory, that incorporate in hardware the ability to learn adapt and evolve, that are designed from the bottom-up to take advantage of the huge benefits, in terms of increases in speed/bandwidth and reduction in power consumption, promised by the emergence of Silicon photonic systems.
The main research efforts will be along the following lines:
- Development of novel computing hardware elements, including neuron and synapse mimics, binary and multilevel memories, arithmetic and computing-in-memory devices.
We will develop basic information processing building blocks (computing primitives) that draw inspiration from biological approaches, providing computing primitives that can mimic the essential features of brain-like synapses and neurons to deliver a new foundation for fast, low-power, functionally-scaled computing based around non-von Neumann approaches.
- Combination of these components together into architectures that deliver the fundamental building blocks of unconventional non-von Neumann processors.
We will combine such computing primitives into reconfigurable integrated processing networks that can implement in hardware novel, intelligent, self-learning and adaptive computational approaches – including spiking neural networks, computing-in-memory and autonomous reservoir computing – and that are capable of addressing complex real-world computational problems in fast, energy-efficient ways.
- Combination of these building blocks into topologies that can address
(i) difficult-to-solve (by conventional means) real-world problems;
(ii) provide localised, intelligent/adaptable, low-power computing nodes.
We will address the application of our novel technologies to future computing imperatives, including the analysis and exploitation of ‘big data’ (e.g. optimisation, correlation, association, pattern determination); and the ubiquity of computing arising from the ‘Internet of Things’.
To realise our goals we bring together a world-leading consortium of industrial and academic researchers whose current work in the development of future information processing and storage technologies defines the state-of-the-art. A list of partners can be found at this link.
We will achieve our aims by combining the remarkable computing capabilities of phase-change devices and nonlinear lasers with novel developments in unconventional computing that can exploit the photonics domain, including neuromorphic processing, reservoir computing and computing-in-memory.
Progress and achievements will be displayed at this link.