Chips that run 1000 times faster than today’s computers could be with us in the next few years, thanks to nanotubes of carbon acting like lifts.
Researchers from Stanford, Carnegie Melon, Berkeley and the University of California detailed plans for a “skyscraper chip” based on carbon nanotube transistors (CNTs). These components stack on top of each other rather than lying them out flat.
The project is called Nano-Engineered Computing Systems Technology, or N3XT, promises to remove bottlenecks by linking processors and memory like floors in a skyscraper and by joining these components with millions of “vias”, which act as electron lifts. The approach would move vastly more bits of data using much less energy.
To enable these advances, the N3XT team uses new nano-materials that allow its designs to do what can’t be done with silicon – build high-rise computer circuits.
“With N3XT the whole is indeed greater than the sum of its parts,” said co-author and Stanford electrical engineering Professor Kunle Olukotun, who is helping optimise how software and hardware interact.
Engineers have tried to stack silicon processors before but with limited success. Making a chip needs temperatures close to 1,800 degrees Fahrenheit, making it extremely challenging to build a silicon chip atop another without damaging the first layer.
The N3XT chips build layers of processors and memory directly on top of one another connected by tiny lifts that move electrons and thus data far quicker. It also incorporates thermal cooling layers to prevent the heat rising from stacked components from degrading performance.
A four-layer prototype of the N3XT chip was demoed at the International Electron Devices Meeting in December 2014, with two layers of RRAM and two transistor layers.
According to Stanford computer scientist Chris Re, the technology could open up vast areas of computing research.
“There are huge volumes of data that sit within our reach and are relevant to some of society’s most pressing problems from health care to climate change, but we lack the computational horsepower to bring this data to light and use it,” said.
“As we all hope in the N3XT project, we may have to boost horsepower to solve some of these pressing challenges.”