IBM’s Neurosynaptic Chip Aims To Transform Science And Technology
IBM deepens its cognitive computing capabilities with new energy-efficient chip the size of a postage stamp
IBM continues to develop its computer chip expertise with a new ‘neurosynaptic’ processor that Big Blue says can mimic the human brain.
This second generation chip is the result of a decade worth of research, and comes after IBM revealed the first hardware prototype in August 2011, which featured a single core.
IBM believes that the second generation of this technology could open up the era of vast neural networks. The new chip, developed in conjunction with Cornell Tech and iniLabs, has a brain-inspired non-von Neumann computer architecture that boasts one million neurons and 256 million synapses.
IBM said that it has built the chip on Samsung’s 28nm process technology. Besides offering 256 million programmable synapses and 46 billion synaptic operations per second per watt, the chip apparently needs a tiny amount of power to operate, despite its 5.4 billion transistors.
Remarkably, it only consumes a minuscule 70mW – that’s several times less power than a modern microprocessor. IBM says that the technology could power a neurosynaptic supercomputer the size of a postage stamp that runs on the energy equivalent of a hearing-aid battery.
This development, according to IBM, “could transform science, technology, business, government, and society by enabling vision, audition, and multi-sensory applications.”
“There is a huge disparity between the human brain’s cognitive capability and ultra-low power consumption when compared to today’s computers,” said IBM. “To bridge the divide, IBM scientists created something that didn’t previously exist – an entirely new neuroscience-inspired scalable and efficient computer architecture that breaks path with the prevailing von Neumann architecture used almost universally since 1946.”
“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 Dr. Dharmendra Modha, IBM fellow and the company’s chief scientist at the Brain-Inspired Computing division of 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,” he added. “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.”
IBM said that the Defense Advanced Research Projects Agency (DARPA) has funded the project since 2008 with approximately $53m (£31.5m) of investment.
“It is an astonishing achievement to leverage a process traditionally used for commercially available, low-power mobile devices to deliver a chip that emulates the human brain by processing extreme amounts of sensory information with very little power,” said Shawn Han, VP of Foundry Marketing ay Samsung Electronics. “This is a huge architectural breakthrough that is essential as the industry moves toward the next-generation cloud and big-data processing.”
IBM has been pushing efforts to drive more intelligence into an increasingly wider range of devices in recent years. It has also sought to create ways to more quickly and intelligently collect, analyse, process and respond to data. These efforts were on public display in early 2011 when IBM’s “Watson” supercomputer beat human contestants on the game show “Jeopardy”.
And IBM continues to invest heavily in its chip research. In July, IBM announced a $3bn (£1.7bn) investment in chips over the next five years, aimed at discovering semiconductor breakthroughs and cramming ever more components onto a piece of silicon.
Yet despite this large investment, Big Blue had been seeking to offload its chip-making business. IBM was reportedly in talks with Globalfoundries about selling its semiconductor division, but late last month, both sides walked away from a deal after failing to agree a price.
Globalfoundries reportedly made an offer that was rejected by IBM as too low. But Globalfoundries had apparently placed little or no value on IBM’s factories because they were too old.
Do you know all about IBM, the founder of the IT industry? Take our quiz!