Posted by Rob Slade on September 19, 2016.
Right. I promised to expand on the security implications of quantum computing.
First off, I probably need to discuss what quantum computing is. Not in a huge amount of technical detail. If you want to get into that, there are lots of papers out on the net for you to read.
Let’s start with bits. If you have any knowledge of computers at all, you know what a bit is. It is the basic particle of information storage. It can be either on or off, representing either a one or a zero; or possibly a true or a false. It can be one or the other, but not both. This is the basis for all modern Turing architecture computers, and the basis of the old joke about how you should never get upset about anything you see on the Internet, since it’s all just ones and zeroes.
Quantum computers don’t have bits. They have qubits. These are quantum entities that partake of a quality called superposition. Unlike bits, which can be a one or a zero, qubits can be one and zero, at the same time. (In some architectures, they can also be anything in between.)
OK, if you truly understand physical computer architecture, at a basic level, that is going to make your brain hurt. But it’s true, and it works. You can have a qubit that starts as both a one and a zero, do a process, and get a definite answer. Trying to go into the “how” would require a lot of understanding of quantum physics and mathematics, so I’m going to leave that out, too. Again, if you want to get into that in detail, there are lots of good papers, and even some courses, online.
Let’s go back to traditional, digital computers. You have a bunch of bits loaded with your initial data. Then you carry out a process, or algorithm, that does a calculation on your initial data. At the end of the algorithm, you get your answer.
However, if you have initial data that has lots of possible options, then you have to load up one possibility, do the process, get an answer, and then load the next possibility, do the process, and so on. You have to reload and run the process for each possibility. (Then you have to compare all the answers you got, and see which one you want.) Even if the process is short, when you have a lot of possibilities the overall proceeding can take a very long time.
However, with a quantum computer, you have qubits. Therefore, you can start out with a situation of “all possibilities.” You can, therefore, run the quantum algorithm once, and come out with the right answer.
Which, for certain types of problems, speeds things up considerably. We’ll look at what types of problems are best suited to quantum computers in a bit.Submitted in: Rob Slade, Uncategorized |