Inspired by the brain’s structure, Scientists collaborating with IBM’s Brain-Inspired Computing group, have developed an efficient, scalable, and flexible Non–von Neumann architecture that leverages contemporary silicon technology and as a result a million computational units could be packed into a tiny patch of area.  It is being touted as a supercomputer the size of a postage stamp.

Putting things into perspective, Dr Dharmendra Modha, the publication’s senior author, notes, “The cumulative total is over 200 person-years of work. Our chip integrates computation, communication and memory very closely.”

The scientists built a 5.4-billion-transistor chip with 4096 neurosynaptic cores interconnected via an intra-chip network that integrates 1 million programmable spiking neurons and 256 million configurable synapses. Chips can be tiled in two dimensions via an interchip communication interface, seamlessly scaling the architecture to a cortex-like sheet of arbitrary size. The architecture is well suited to a variety of applications that use complex neural networks in real time, for example, multi-object detection and classification. With 400-pixel-by-240-pixel video input at 30 frames per second, the chip consumes 63 milliwatt which amounts to very little energy.

Dr. Modha believes that the processor is “a new machine for a new era”. However, plenty needs to be accomplished before its time for the chip, dubbed TrueNorth, to be commercially useful.

The delay is partly owed to the fact that this chip will require tailor-made programs as opposed to the traditional style which was conceived in the 1940s and still powers nearly all modern computers. The design, where the processors and memory are separate, tends to be a logical match for sequential, mathematical operations but, the thoroughly interconnected structure of biologically-inspired, “neuromorphic” systems like TrueNorth is said to be a much more efficient way of handling a lot of data at the same time.

Read more here

(Image Credit: Saad Faruque)

Previous post

6 Stumbling Blocks for Marketing With Data

Next post

Is Your Big Data Vision Focused on Designing Everyday Apps?