Supercomputers are used today for various purposes, used in machine learning, and can even provide accurate weather prediction. However, they need high – really high – processing power to accomplish their tasks, and some companies are always dedicated to investing in this type of computing. An example of this is IBM, which announced a partnership with the US Air Force for yet another super machine.
Famous for toasting Watson, a computer dedicated to artificial intelligence, IBM is teaming with the Air Force to build the world’s first supercomputer that behaves like a natural brain.
IBM and USAF announced on Friday that the machine will run on a series of 64 TrueNeorth Neurosynaptic chips. TrueNorth chips are connected together, and should operate similarly to synapses within a biological brain. Each core is part of a distributed network and they operate in parallel with each other on an event-driven basis.
With this, these chips do not require clocking, like conventional CPUs, to work. Also, due to the distributed nature of the system, even if one core fails, the rest of the array will continue to function. This 64-chip array will contain processing equivalent to 64 million neurons and 16 billion synapses – but each processor consumes only 10 watts of electricity.
Like other neural networks, this system will be used in functions of pattern recognition and sensory processing. The Air Force wants to use TrueNorth’s ability to convert various types of data – whether audio, video or text – into machine-readable symbols.
“AFRL was the earliest adopter of TrueNorth for converting data into decisions,” said Daniel S. Goddard, director, information directorate, U.S. Air Force Research Lab. “The new neurosynaptic system will be used to enable new computing capabilities important to AFRL’s mission to explore, prototype and demonstrate high-impact, game-changing technologies that enable the Air Force and the nation to maintain its superior technical advantage.”
“The evolution of the IBM TrueNorth Neurosynaptic System is a solid proof point in our quest to lead the industry in AI hardware innovation,” said Dharmendra S. Modha, IBM Fellow, chief scientist, brain-inspired computing, IBM Research – Almaden. “Over the last six years, IBM has expanded the number of neurons per system from 256 to more than 64 million – an 800 percent annual increase over six years.”