IBM's New Chip Has Brain-Inspired Computer Architecture

Claiming a big step toward cognitive computing, IBM unveiled a chip called SyNAPSE that uses a brain-inspired non-von Neumann computer architecture. The von Neumann architecture has been in use almost universally since 1946. The neurosynaptic computer is the size of a postage stamp and runs on the amount of energy equivalent to a hearing aid battery.

The company says this technology could transform science, technology, business, government and society at large. This is the first neurosynaptic computer chip to achieve the scale of one million programmable neurons, 256 million programmable synapses and 46 billion synaptic operations per second per watt.

The 5.4 billion transistors chip is built on Samsung’s 28nm process technology. The new chip bridges the divide between the human brain’s cognitive capability and ultra-low power consumption. It is one of the largest CMOS chips ever built but consumes a meager 70 milliwatts during real-time operation, which is far less than a modern processor.

The second-generation chip is the culmination of almost a decade of research and development. It brings us a potential future of neurosynaptic supercomputers. A single-core hardware prototype was unveiled in 2011. In 2013, a new software ecosystem with new programming language and chip simulator was released.

The chip's brain mimicking design means it can enable vision, audition, and multi-sensory applications.

The new cognitive chip architecture has an on-chip two-dimensional mesh network of 4,096 digital, distributed neurosynaptic cores. Each core module integrates memory, computation and communication and operates in an event-driven, parallel, fault-tolerant fashion.

To enable system scaling beyond single-chip boundaries, adjacent chips, when tiled, can seamlessly connect to each other, building a foundation for future neurosynaptic supercomputers. To demonstrate scalability, IBM also revealed a 16-chip system with sixteen million programmable neurons and four billion programmable synapses.

“IBM has broken new ground in the field of brain-inspired computers, in terms of a radically new architecture, unprecedented scale, unparalleled power-area-speed efficiency, boundless scalability and innovative design techniques,” said Dharmendra Modha, IBM fellow and chief scientist for brain-inspired computing at IBM Research. “We foresee new generations of information technology systems – that complement today’s von Neumann machines – powered by an evolving ecosystem of systems, software and services.

"These brain-inspired chips could transform mobility, via sensory and intelligent applications that can fit in the palm of your hand but without the need for Wi-Fi.”

Note: Article originally incorrectly stated kilowatts instead of milliwatts.