IBM has come up with a computer chip that closely resembles the human brain in its functioning and takes computing capability to a new level.
With a total of 5.4 billion transistors the computer chip, named TrueNorth, is one of the largest metal oxide semiconductor chips ever built.
Working together with traditional computing devices, the chip can provide a dose of human intelligence for smart car sensors, cloud computing applications or smartphones. This includes processing information in real-world environments like recognizing a sound or a face in a moving crowd.
Derived from the way the human brain functions with a billion neurons connected with thousands of others and processing information in parallel, the IBM chip consists of one million programmable "neurons" and 256 million programmable "synapses" conveying signals between the digital neurons.
Each of the chip's synaptic cores handles memory, computation, and communication, operating in parallel based on "event-driven" computing.
Traditional computing does not allow for program instructions and operation data to pass through the same route simultaneously.
The human brain has around 100 billion nerve cells or neurons and each connected to some 10,000 others, making for about 1000 trillion connections.
"Our long-term end goal is to build a 'brain in a box' with 100 billion synapses consuming 1 kilowatt of power," Dharmendra Modha, chief scientist and founder of IBM's Cognitive Computing group at IBM Research-Almaden says.
"In the near future, we'll be looking at multiple things for empowering smartphones, mobile devices and cloud services with this technology."
Despite its size, the chip uses just 70 milliwatts while running and has a power density of 20 milliwatts per square centimetre — almost 1/10,000th the power of most modern microprocessors. This brings its efficiency much closer to the human brain's astounding power consumption of just 20 watts.
Traditional computer chips have a clock that uses power to trigger and coordinate all the computational processes. But the IBM chip's digital neurons can work together asynchronously when triggered by the signal spikes, thus reducing the power demand.
The other advantage is that the chip architecture allows for easy scalability. The chips themselves can be arranged in a repeatable pattern to create bigger machines. IBM has already tested that idea with a 16-chip configuration.
The chip's architecture was detailed in a paper published in the 7 August online issue of the journal Science.