From the labs: IT's future today

The future of IT is taking shape in the form of next-generation computing technologies under development in today's R&D labs

For all its promise of revolution, the computing industry often lags behind expectations. After all, your netbook is really just a laptop, only smaller and cheaper. The chip that powers your PC today has a direct lineage back to the Pentiums of yesterday. Your latest hard drive might hold 2TB, but it's still just a hard drive. Where's the real innovation?

In the labs, of course.

[ Relive 15 turning points in tech history and find out which technologies made InfoWorld's list of all-time tech flops. | Weigh in on 12 crackpot tech ideas that could transform the enterprise and find out which emerging enterprise technologies will have the biggest impact in the years to come. ]

Researchers in labs at major IT vendors and universities continue to point the way forward. Products and ideas dreamed up in these labs have the potential to shake up the IT industry. From the network to storage systems to the securing of sensitive data to the way in which end-users will one day interact with computing interfaces -- every facet of the enterprise is being pushed in revolutionary directions.

Off the Record submissions

Here are just a few of the ideas evolving in today's labs -- future technologies that could be arriving sooner than you think.

Processors: Breaking Moore's Law
The story of computing since the mid-20th century has largely been the race against Moore's Law. Named for Intel co-founder Gordon E. Moore, the conjecture posits that integrated circuits will double in performance every 18 months. In practice, Moore's prediction has held mostly true. Even as chipmakers have drawn close to the practical limits of modern processor design, the advent of multicore CPUs has allowed them to cram ever more power into increasingly compact chip packages.

But some scientists see Moore's Law nearing the end of its useful life, and not because processors will stop getting faster. Rather, they believe computing will soon undergo a quantum leap. Quantum computers, devices that derive their processing power from the curious properties of subatomic particles, are widely seen as the next major evolution in computing. But while scientists struggle to build practical quantum computers -- even the most successful experiments to date have solved only simple calculations -- a number of other processor technologies in the works could prove equally revolutionary.

At the University of Illinois, Professor Rakesh Kumar has proposed a novel way to improve the speed of current CPUs: Cut the brakes. Today's chips, he says, spend too much time trying to get every calculation exactly right, while real-life integrated circuits make mistakes all the time. Ensuring flawless performance forces chips to consume more power than necessary, making it difficult to shrink the size of the components on the chip. Not every crash is a catastrophe, says Kumar, whose team is working on fault-tolerant rather than error-free processor designs. Combined with software that can cope with CPU errors when they occur, these so-called stochastic chips could run faster at lower power levels without worrying about flying out of control.

Meanwhile, teams from Michigan Technological University and Japan's National Institute for Materials Science are trying a far more radical approach. They've forgone silicon altogether, instead using DDQ -- a compound composed of carbon, nitrogen, oxygen, and chlorine -- to build a molecular computer that they claim already mimics the parallel processing structure of the human brain. A truly brainlike computer would outperform current CPUs by orders of magnitude. And because their design uses organic molecules, the researchers believe such computers could be grown from algae, rather than having to be constructed using industrial chemical processes.

1 2 3 4 Page 1
From CIO: 8 Free Online Courses to Grow Your Tech Skills
Notice to our Readers
We're now using social media to take your comments and feedback. Learn more about this here.