Most everyone who has used a computer at one time or another has had to deal with the unpleasantness of that computer crashing. For the most part, it's a minor inconvenience that calls for a quick reboot. Other times it takes important projects with it because it crashes before it can save. But the days of computers crashing may be long gone thanks to a recent discovery at University College London, which led to the development of a whole new kind of computer that will never crash.
So how does this little miracle box work? It's actually fairly simple. The computer will automatically repair any corrupted data that it finds whenever that data shows up, meaning that anything that might cause a computer to crash is spotted, intercepted, and repaired before it can impact the operations of the rest of the system.
As explained by Paul Marks of "New Scientist," computers constantly work through a list of instructions, which are pulled from memory, and then the results of those instructions are also stored in memory. When something goes wrong in the process, crashes happen, since the computer doesn't really know what to do next. It then provides an error message telling those who are particularly savvy just what went wrong, while everyone else just grumbles and restarts. But with the new self-repairing computer, things work a bit differently.
Instead of running off one big list, it divides that list into a set of smaller digital slots. These slots, called "systems," are then executed randomly by the selection of a random number, and instructions are carried out all at once. The random number is actually selected by what's called a "pseudorandom number generator," which isn't completely random, but is close enough to random that it works in this case. This way, when there's an unexpected problem, the systems can carry on around that particular obstruction to generate the necessary results despite the problem in the overall environment as no one system comes before another.
While this bodes well for computer users, the device's inventors believe that it can be used elsewhere too, like in reprogramming drone systems automatically, or in providing a better model of the human brain, as that particular organ can also work around blockages in a similar method. That last advance is likely of particular interest as the brain is still an organ that's largely poorly understood. Of course, computers that can repair themselves as needed will likely mean a bit of a drain on the IT industry, but considering the variety of tasks that the department executes, layoffs as a result of self-repairing computers are likely to be small.
Still, the University College London team will be taking its findings to Singapore this April for exhibition, and we may well start seeing this technology in place before too much longer has passed. The idea of ditching the blue screen of death, or its Apple (News - Alert) equivalent, is likely a development that many users will cheerfully welcome.