Podcast: Quantum Computing Takes Flight

Traditional computing has relied on rapid development to meet performance goals. However, advancement has slowed as circuits have gotten smaller, and many see a need for revolutionary new hardware architecture to continue driving computing into the future.

In a recent podcast interview, Financial Sense spoke with Vern Brownell, CEO of D-Wave Systems, about their quantum computers, how they work, the companies using them, and why, he believes, we are at the "dawn of the quantum computing age."

(Subscribers can access the full audio by clicking here or via a podcast app on their mobile device. Not a subscriber? Click here.)

First, to start off with the basics.

What is Quantum Computing?

A quantum computer is just a computer that uses the laws of quantum mechanics to do computations, Brownell stated. The physics community has theorized that such computers would be much more powerful than classical models because they use qubits instead of bits.

Bits use 1s and 0s as the base units to perform calculations, where qubits can be either 1s, 0s, or both 1s and 0s simultaneously, Brownell said.

“The special properties of these qubits are really at the core of all quantum computers,” he said. “They can operate in both states at the same time, which is called superposition. That’s a fairly unique property.”

The company’s computer incorporates 1000 qubits connected together, which enables the machine to provide an enormous number of answers at the same time, around 10 to the 130th power, Brownell said. To put that in perspective, scientists estimate there are only 10 to the 90th atoms in the universe, noted Brownell.

“That’s just an example of the enormous scalability that comes from quantum computing in general,” he said.

Who’s Involved?

D-Wave counts both NASA and Google as customers, which have collaborated to create the Quantum Artificial Intelligence Laboratory (QuAIL) in Mountain View, California, and are looking at the use of this technology for machine learning applications primarily, Brownell noted.

“Their belief is that this type of computing power can be beneficial to all of the machine learning that goes on at places like Google,” he said. “Google is probably one of the world’s masters in machine learning, and it drives almost every part of their business.”

Lockheed Martin has also purchased one of the company’s machines, Brownell noted. For Lockheed and other defense contractors, quantum computing has a wide range of applications, including mission planning and logistics, systems validation and verification, as well as pattern recognition and anomaly detection.

Why Quantum Annealing?

There are likely six or so different ways of creating quantum computers, Brownell says. D-Wave chose to use quantum annealing primarily because the company thought it would be the most practical way to implement the technology.

“That’s turned out to be very true,” he said.

Quantum annealing takes a piece of matter – in this case the computer’s chip – and turns it into a computational device that does an optimization, Brownell said. Optimization problems are prevalent in computer science and basically involve trying to find the best answer out of a range of many different answers, he added.

Applications include machine learning, financial forecasting and risk models, graph theory problems, pattern recognition, advanced search, and many other areas.

“In financial services we see things like Monte Carlo simulations and portfolio optimization, which are examples of applications that fit well on quantum annealing,” he said.

How Fast Is It?

At the end of last year, Google announced that they had identified a problem where the D-Wave computer performed 100 million times faster than an individual Intel core, Brownell said.

Though the particular problem is what is known as a benchmark problem, which is designed to test a system’s limits and isn’t necessarily an indication of how the machine will perform in a given task, it does highlight the capability of the machine, Brownell stated.

That performance speed is eight orders of magnitude faster for this particular problem compared to traditional computing methods, he said.

“It’s a very powerful empirical proof that this is quantum mechanics that is giving the computational advantage,” Brownell stated. “The gentleman we work with at Google made the analogy … of this being like the Wright brothers’ first flight—the first time that quantum computing has clearly been demonstrated to outperform classical computing in history.”

How Is It Implemented?

The company’s computer essentially resembles a 10-cubic-foot black box, Brownell said, but the actual processor performing the calculations is the size of a person’s fingernail.

The chip operates at a very low temperature—minus 273.15 degrees Celsius or 180x colder than interstellar space—which is close to absolute zero, and also has be shielded from outside interference such as magnetic fields.

“All the bulk of this machine is really just to create this very special protective environment,” he stated.

Because of the extreme requirements to operate effectively, the chip dissipates no heat, Brownell noted, and would primarily be put to use in data centers, where its relatively low power consumption would be a boon.

He believes quantum computing will be used alongside classical computing, and not serve as a replacement. This hybrid setup will allow cloud providers to offer quantum-based computing solutions to anyone in the world, he said.

Does It Work?

Some have questioned whether the company’s machine is actually using quantum processes. Quantum computing of any kind is quite complex, Brownell noted, and the company’s choice to use quantum annealing is part of the controversy.

Though other types of quantum computing have been proposed, the company decided these would be very hard to implement, and instead chose to go with quantum annealing.

“So now we’re in this great position where we’ve built this machine, we’re roughly a decade ahead of anyone in the field, and there’s been a lot more validation of quantum annealing,” Brownell said. “Our computer is not a so-called ‘universal’ quantum computer, but that says nothing about its practical usefulness.”

“Between the 60 peer reviewed papers and work that Google just did, it’s very hard to deny that quantum mechanics are at the core of what we do,” he said. “We feel very strongly that this is the dawn of the quantum computing age.”

To listen to this entire podcast interview with Vern Brownell, CEO of D-Wave Systems, log in and click here. Not a subscriber? Click here.

About the Author

fswebmaster [at] financialsense [dot] com ()
randomness