Julie Grollier
spin torque

bio-inspired architectures are very relevant today

The CMOS industry is facing a number of hurdles to continue improving processors in terms of size, performance and dissipation. While the number of transistors keeps increasing, since 2005, their frequency is stalling due to the extremely large thermal dissipation. In order to keep on increasing performances, computer architects have introduced parallelism in their systems via multi/many-core architectures. Nevertheless other technical complications will have to be solved such as dark silicon and the increasing number of defective components due to their shrinking dimensions.

In this context, bio-inspired computing architectures are seen by many, as a pertinent solution to complement classical processors Indeed, bio-inspired architectures are massively parallel, large scale, and highly interconnected. This confers them extremely interesting qualities such as speed, low energy consumption and high tolerance to component defects and variability. The best example is our our brain, which can interpret complex scenes in fractions of seconds, demands only a few meals a day (less than 10 MJ, to be compared to 105 MJ for a supercomputer) and manages to compute despite the fact that we lose about 104 neurons a day.

In addition bio-inspired architectures such as artificial neural networks are very well adapted to the Recognition, Mining and Synthesis emerging applications required to deal with the current data explosion.