Breakthroughs in computing are enhancing our ability to make sense of large bodies of data, providing guidance in some of the world’s most important decisions, and potentially revolutionizing entire industries.
The term ‘cognitive computing’ refers to systems that, rather than being explicitly programmed, are built to learn from their experiences. By extracting useful information from unstructured data, these systems accelerate the information age, helping their users with a broad range of tasks, from identifying unique market opportunities to discovering new treatments for diseases to crafting creative solutions for cities, companies, and communities.
With systems growing in size and complexity, traditional computer architecture seems to be reaching its limits, as power consumption soars and the transmission delay between components becomes increasingly burdensome. Rather than attempting to squeeze energy-intensive performance out of ever-larger chips, scientists at IBM are experimenting with arranging computer components in a dense 3D matrix that maximizes not performance, but energy efficiency.
Arranging computer chips in a 3D environment puts the various elements of the computer closer to one another. This not only reduces the time they take to communicate; it improves energy efficiency by a factor of as much as 5,000 – potentially providing computers with efficiency close to that of a biological brain. Already, a much denser computer built from available mobile technology and hot water cooling allows for ten times higher efficiency than a conventional system.
But man-made computers are so inefficient not only because they need to power the chips, but also because they need energy to run the air conditioners that remove the heat generated by the processors. A computer on the 3D model could use coolant fluid to deliver energy to the chips. In addition to dissipating heat, the fluid could be used to power an electrochemical system providing power to the processors. This, in turn, would allow for further increases in packaging density – and thus efficiency.
By adopting some of the characteristics of the human brain, computers have the potential to become far more compact, efficient, and powerful, allowing us to take full advantage of cognitive computing – providing our real brains with new sources of support, stimulus, and inspiration.