Quantum Computing 101, with D-Wave's Vern Brownell | Big Think

Quantum computing is a whole new category of computing and it directly leverages the laws of quantum mechanics to do a computation. As we all know, quantum mechanics are the most fundamental laws in the universe. It describes how everything in the universe works.

So what we've built and what other quantum computing researchers have done is create computers that directly use those laws of quantum mechanics. And that sounds fairly straightforward but, in fact, it's quite difficult to do because the enemy of quantum computing is the environment. And when I say the environment, I mean things like temperature.

And when you have temperature, you have molecules moving around that cause interference to the quantum computation. You also have electromagnetic interference from radio sources and gamma rays and all sorts of things. So you need to create a very quiet, clean, cold environment for these chips to work in.

And ultimately, what we're building is a quantum computer on a chip that's about the size of your fingernail in this very exotic environment. So that environment runs at near absolute zero. Absolute zero, as you know, is the lowest temperature possible in the universe. It's also called zero degrees Kelvin.

So these machines run at a very low temperature so that they can have that pristine, very clean, quiet environment to run in and it doesn't disturb that quantum computation. And, in fact, it runs down at what's called 10 millikelvin, which is 0.01 Kelvin. Absolute zero is zero degrees Kelvin, so this is running at minus 273.14 degrees C and the lowest possible temperature in physics is minus 273.15 degrees C.

So very, very cold. A very, very rarified environment because we're also running in effectively a magnetic vacuum. So you could consider these environments, these rigs that we built, these systems that we built to be probably the most rarified environments in the universe unless there's other intelligent life in the universe that has, you know, pure colder environments.

For instance, outer space is 150 times warmer than the environment that we built for these quantum computations. So you may ask why do we go through all this trouble? The answer is the problems of quantum computing is exponential speed ups over classical computing for a particular set of problems.

And that's very important and exciting to researchers that are working on that kind of human scale problem ranging from things like developing drugs for cancer or better modeling the molecular interactions of cancer and how it attacks cells and things like that to big data analysis, looking for patterns and inferences and drawing insight from large amounts of data or doing things like better modeling financial services markets and better managing risk and so on.

So there's all kind of applications that aren't particularly well suited by today's type of computers and I refer to today's computers as classical computers. They compute largely in the same way they have for the past 60 or 70 years since John von Neumann and others invented the first electronic computers back in the 40s.

And we've had amazing progress over those years. Think of all the developments there have been in the hardware side and the software side over those 60 or 70 years and how much energy has been put -- energy and development has been put into those areas. And we've achieved marvelous things with that classical computing environment.

But it has its limits too and people sometimes ask why would we need any more powerful computers. These applications, these problems that we're trying to solve are incredibly hard problems and aren't well suited for the architecture of classical computing.

So I see quantum computing as another set of tools, another resource, set of resources for scientists, researchers, computer scientists, programmers to develop and enhance some of these capabilities to really change the world in a much better way than we're able to today with classical computing.

It's not a replacement for classical computing. It will be used in what I would call a hybrid approach where you're going to see both the capability that's already been built in high performance computing and other types of computing markets working very closely with quantum computing resources.

One of the fundamental building blocks or the fundamental building block of a quantum computer is a thing called a qubit -- Q -- U -- B -- I -- T, right. So it's basically a bit which is the lowest level building block of a computer today. It's either a zero or a one in digital terms today with classical computing.

A qubit has the interesting property that it can be in zero and one as a digital bit can be today but it can also be in what's called the superposition of zero and one. In other words, it can be in two states, zero and one, at the same time.

And our minds have kind of difficulty understanding that. How can a particle or an object or a qubit -- how can it actually be in two states at the same time? You know, it's just a very foreign concept to us but it's been proven over and over. It's a well understood characteristic of quantum systems that they operate in superposition and entanglement is another quantum mechanical property that's well understood.

A little bit spooky but certainly well understood. So these qubits are interesting and you might think, well, okay, that's interesting. I can have zero, one and the superposition of zero and one at the same time. But it only really starts to get interesting when you can string qubits together.

So if you put say 512 of these qubits together as we have in our latest generation processor, you can then represent two to the number of qubits states simultaneously. So at the very beginning of the computation in our computer we're actually in two to the 512 different states at the same time.

So two to the 512 is ten to the 154th power. There's probably only, scientists estimate, ten to the 80th atoms in the universe. So it's an absolutely astounding number. And it's indicative of the kind of amazing things that go on regardless of what type of quantum computer you have when you have these qubits and you can implement them and string them together and couple them and entangle them.

You see that as we grow the number of qubits that capability grows very dramatically and but that's not -- a caveat there. That analogy is not directly related to the performance that you'll see from this machine over time because computing performance has a lot of other characteristics that go along with it, namely the precision that you can specify, the problem, the coupling between qubits and all sorts of other things.

So it's sort of like, you know, Intel or others say their integrated circuits have many billions of transistors. That's not a direct implication to performance. It's kind of a metric or rule of thumb, you know, the more transistors you have the more capability but it's not directly related to performance.

So qubit is that fundamental building block that's the basis of all quantum computers. So there are different ways to build quantum computers. And I would actually say that quantum computing is really a category of computing because there will be many different types of quantum computers that will be built over the next decades.

There's basically the theory in how you would build a quantum computer and then there's the implementation of how you do it. And those are sometimes two separate things. Our computer uses a theory that was developed at MIT in roughly the year 2000 by Ed Farhi and a team of researchers in the physics department at MIT.

It's called adiabatic quantum computing. It uses this quantum annealing capability. And our founders believed that it was much easier to implement this type of quantum computing. It was also more robust against noise, some of those things that I talked about trying to protect the environment from the noise and the interference of the environment.

That's really turned out to be true. The other types of quantum computing -- the predominant one that people are trying to implement today is called gate model quantum computing which is an entirely sensible way of doing things because what you try to do is to replicate digital gates which are the building blocks of all computers today and build quantum equivalents for those gates.

The problem with gate model computing is it's exceptionally difficult to build and there's a property called decoherence meaning basically the interference from the outside environment causes the computation to be shortened or short lived and it causes a lot of difficulty in building large scale quantum computers.

So the largest gate model quantum computing effort that's been done today, I think, is factoring the number 21, seven and three as we all know. So it's a very small scale kind of experiment kind of thing. And that really hasn't improved dramatically over the last decade where D-Wave with its choice of adiabatic quantum computing has been very effective at solving real world problems at real scale.

There was another choice the D-Wave made that was equally as important in how you implement the type of quantum computing. So there are different ways of actually building kind of the substrate of quantum computing and that can range from manipulating ions or atomic particles at a very atomic scale.

And obviously that can be difficult but typically the way people do that is by having lasers that energize a particle and manipulate the particle. In fact, Dave Wineland, who just won the -- last year I believe, won the Nobel Prize in physics for his work that was done with ion trap quantum computing.

The method that we chose is called superconducting electronics and, as you may know, superconducting electronics are very low temperature electronics that exhibit this really interesting property at very low levels. Resistance goes to zero. They start to superconduct and there are two reasons why we use superconducting electronics.

One is we need it to run at that low temperature anyway to minimize the interference from the environment. But also, this superconducting technique has an interesting side effect that's important that it generates no heat. So as these computers scale, as our computers scale, because we've implemented superconducting, the actual heat dissipation of the chip doesn't scale which is completely contrary to all of computing that we know today that it becomes a huge energy drain.

In fact, one of the largest consumers of electricity in the world now is the computing environment because of this dissipation or this resistance that's in primarily CMOS is the technology that's used to build computers today. So an interesting side effect of what we built is it will be very energy efficient, in fact, not using any power at all other than the power it takes to drive the refrigerator and the equipment that goes along with that.

But that'll be a relatively modest fixed heat load. So there are other -- there are even other types of quantum computers that people are building. Microsoft has a really interesting project going on where they're trying to develop what's called a topological quantum computer.

It's yet another type of quantum computer and that is very interesting and some scientists, computer scientists are very excited about the potential there. The problem is that type of computing will require the discovery of a particle called a non-abelian anyon which physicists do believe exists but they haven't actually been able to identify one.

So once they identify that particle, then they can start to think about how they would build the hardware to harness that particle and so on. So that's a, you know, probably a very long term effort in order to build something like that -- maybe decades.

So the inspiration of our founders were really let's get to market with something that can deliver real benefit to the computing users of the world as soon as possible. And they made that decision more than ten years ago and we've been implementing that ever since.

And I think that's put us in a good position and that's why we've been able to deliver to our customers and partners today.