One nanotechnology researcher said supercomputers small enough to fit into the palm of your hand are only 10 or 15 years away.
"If things continue to go the way they have been in the past few decades, then it's 10 years," said Michael Zaiser, a professor and researcher at the University of Edinburgh School of Engineering and Electronics. "The human brain is very good at working on microprocessor problems, so I think we are close - 10 years, maybe 15."
Zaiser's research into nanowires should help move that timeline along.
For the last five years, he has been studying how tiny wires - 1,000 times thinner than a human hair - behave when manipulated. He explained that each such miniscule wire tends to behave differently when put under the same amount of pressure. Therefore, it has been impossible to line them up close to each other in tiny microprocessors in a production atmosphere.
Zaiser said he's now figured out how to make the wires behave uniformly. He separates the interior material of the wire into distinct groups so the wire can't react as a whole. That makes it much easier to control. "It's like crowd control," he added. "If they can all go one way, you have a big mess."
These nanowires will go inside microprocessors that could, in turn, go inside PCs, laptops, cell phones or even supercomputers. And the smaller the wires, the smaller the chip can be.
Shrinking down the microprocessors is a big step toward shrinking down computers.
Zaiser was quick to point out that his nanowire discovery won't immediately lead to the development of supercomputers that can fit in the palm of a hand or even shrink down to the size of a book of matches. In addition to smaller microprocessors, engineers will need to deal with thermal fluctuations that erupt at that size.
But he does humbly admit that taming nanowires is a big step toward the goal of small supercomputers.
"This will enable chips to become much smaller," he said. "Think 10 years back. You could hold [a cell phone] the size of a telephone receiver and it didn't work so well. Today, you can fit what is a powerful computer onto a small device."
Charles King, an analyst at Pund-IT in California, said with advances like nanowire technology continuing to come along, Zaiser's predictions for tiny supercomputers may not be so far off. And that will be a huge step for the industry, considering that not so long ago supercomputers filled up enormous rooms or even entire buildings.
"Actually, what he's saying is not crazy," said King. "The wire problem was an important one. Solving that particular issue, one so fundamental, needed to be done. If he can solve that, a lot of people, a lot of companies, will appreciate that. The industry is aligning in a direction where we're going to be seeing continuing improvements and developments."
Jim McGregor, an analyst at In-Stat in Arizona, said he definitely can see palm-size supercomputers coming out within 10 to 15 years, but he thinks they'll be based on the basic silicon technology used today. He added that the next generation of tiny supercomputers - made with carbon nanotubes and nanowires - may be as far off as another 30 to 40 years, though.
"You know, everybody says we'll get to a physical limit but every time we think we're going to hit it, we overcome it," said McGregor. "They end up being speed bumps instead of road blocks."
King added that over the last 10 or 20 years, technology advancements have put smaller and more powerful computers into the hands of children and researchers alike. That leads him to believe that the shrinking of supercomputers is just ahead of us.
"We're at the beginning of some very exciting times in supercomputing," he said.