Towards the end of 2001, IBM made a breakthrough – they proved that three times five is fifteen. Not a big deal, one would say. But they haven’t heard the whole story yet: IBM did it using only seven atoms. That’s right – seven atoms. Not a full-fledged computer with a chip and input mechanism and trillions of atoms making all the components up. Only seven atoms did the trick. That, in short, is Quantum Computing – harnessing the power of nature.
(image source)
Moore’s Law coming to an end
(image source)
Moore’s famous law – computing power doubles every eighteen months. So by that logic, computers of 2050 will be able to process data at the speed of human thought – five hundred trillion bytes per second. This is enough to build machines with AI comparable to human intelligence. But there’s only one problem – Moore’s law won’t stand up till 2050. By 2020, we will have probably expanded it. There are only so many transistors we can fit on a chip. Eventually the distance between them will be too small for us to make any further improvements.
So where do we go once Moore’s Law comes to an end? IBM has recently been showing off Graphene chips. Remember that Moore’s Law corresponds to the use of Silicon. Graphene is a special form of Graphite that consists of a single layer of carbon atoms. IBM managed to reach speeds of 100 GHz using it last year.
Graphene chips are out?
(image source)
But then last week, IBM dismissed the idea of replacing silicon in computers with graphene. They said that because of its small width, graphene does not have a power gap and hence cannot be completely switched off. So while it works well in a laboratory, it is unlikely to replace the transistors in a CPU. While they may complement silicon in hybrid circuits such as RF circuits, hopes of using them to continue Moore’s Law have been well and truly dashed.
What do we need Quantum Computing?
Everything mentioned above leads us to this – Quantum Computing. When Moore’s law comes to an end, we will no longer be worried about improving silicon. Hopefully, by then we will have moved onto something much better – using atoms for computing.
Suppose you want to simulate the behavior of a handful of atoms. With today’s technology, you would need nothing short of a super computer and a couple of days to perform the simulation. Yet in real life, nature can simulate their behavior using just that handful of atoms. That is something we plan to achieve using Quantum Computing.
Quantum Computing will increase computing power not linearly, but exponentially.
Potential uses
Why would we need so much power? For everything! Imagine having the power of the current breed of super computers on your desktop.
“But will it run Crysis?”
Oh! Yes it will! It will run thousands of instances of Crysis at the same time, if you so wish. But desktop implementation is far away.
Think of the super computers locked away in laboratories. If we replace the silicon in them, imagine how much better weather prediction will get. So will tracing the motion of stars and planets, predicting future natural disasters, simulation of events like earthquakes, etc.
Things that require days to compute today (like DNA sequences) will be completed in a matter of minutes.
Just like we didn’t know what computers would be capable of when we first invented them, we don’t know today what we might use Quantum computers in the future for. We might use them to create virtual worlds like the Matrix (imagine how you would be playing Second Life then!). We might use them to create accurate simulations. We might use them for mind reading. We might even use them for teleportation! Because we haven’t truly harnessed the power of Quantum Computing yet, we don’t know what it is capable of.
Security issues
All the uses are not without potential security issues, however. When IBM computer 3*5 using seven atoms, even the CIA took note. If and when quantum computers become a reality, any code the CIA makes will be cracked in a matter of minutes, making password protection and encryption obsolete.
Fortunately, the same computers that can be used to break codes can also be used to make codes. However, this still wouldn’t work unless Quantum Computing made its way into homes. Brute force attacks will become so simple that we will have to completely rethink security. Passwords will no longer work.
Current state
This article from October 2009 outlines the current state of technology. While the technology is exciting, it is so complex that we don’t yet fully understand how it works. As a result, we are not sure how to extract the most out of it. However, with a major focus on the area today, scientists are figuring out quantum computing at a much faster pace than ever before.
A few days back, an international team of scientists created 10 billion qubits on silicon at once. A qubit is an entangled pair of atoms, such that changing the state of one instantly affects the state of another, no matter how separated they are in space.
One of the biggest problems with Quantum Computing, however, is getting the results out. The calculations are performed and stored in qubits. How do you extract the solutions from there? For this, we currently have to go back to our old methods of using silicon, eating up the time saved by using Quantum Computing in the first place.
Another huge problem is that they are easily affected by external disturbances. Even cosmic radiation can throw your calculations off track, since they can cause atoms to vibrate strangely. How to isolate quantum computers from noisiness is a big issue. Isolation is also needed because quantum information has a tendency to leak into the outside environment.
As things stand right now, Quantum Computing is still a far distance into the future. We might be lucky if we get to use them in our life time. But no matter how long it takes, one thing’s for sure- Quantum Computing is well on its way. And it is going to change the world forever.