26th September 2016

Computer Scientist Dr Christian Fensch has been awarded funding from the EPSRC for his research into Approximate Computing Technology, a recently proposed technology aiming to greatly increase energy efficiency.

Computers have revolutionised our lives, from the mobile phones we use on a daily basis to supercomputers which help in the discovery of medical breakthroughs or enable the design of more energy efficient cars and buildings. This progress has been made possible by continuously increasing computational power but, as Dr Fensch explains, we now face two threats to this trend.

“Harnessing this resource has become increasingly difficult. Imagine a car that provides direct control of fuel mix, 20 gears and adjustable valve timing. This car will provide excellent performance, but requires a driver with an engineering degree to make the optimal adjustments.

There is increasing demand for improved computational energy efficiency. We cannot attach larger batteries to a mobile phone, or build a nuclear power station next to each data centre. While there are ongoing discoveries that improve efficiency, these solutions intensify the first problem: they increase the difficulty.”

For a solution to be truly practical it needs to be usable by non-experts
Dr Christian Fensch

The solution posed by Dr Fensch looks at approximate computing providing a ‘good enough' solution rather than the precise and exact solution in traditional computing.

“Precision, when not necessary, is a waste of energy. Imagine quickly looking into your wallet to check how much cash you carry; is it £5, £20 or more than £50? It would be a waste of time to count the cash precisely to find out if you have £17.52 or £17.53. Research has shown that a vast body of problems can take advantage of this kind of imprecision.”

One of the key aims of the project is to make the technology available to non-expert programmers. Worth over £260,000, the project will begin early next year in partnership with Codeplay Software Ltd and the University of Washington.