A quantum leap for computing

Originally published in The Australian.

Able to calculate problems at an astronomical scale, quantum computing is getting closer to reality.

For years technologists have questioned how much longer we might expect the current rate of advancement in computing continue.

While Moore’s Law has promised that the speed and capability of computers will double every two years, this has meant components have had to be made smaller and smaller. Barriers loom however in the form of manufacturers’ abilities to work at these microscopic scales.

But while making components smaller can be problematic, the properties of physics that operate at the atomic scale also holds the key to unleashing computing power much greater than Moore’s Law has ever promised.

Called quantum computing, this next generation of computers take advantage of a set of properties called quantum mechanics, and their specific ability to allow particles to exist in multiple simultaneous states, called superposition.

According to the vice president and head of exploratory science and university partnerships at IBM Research, Jeff Welser, quantum computing can be thought of as an extension beyond what we can do with a classical computer.

Classical systems represent information as a ‘bit’, where its state is either one or zero (the foundation for binary computing), while in quantum computing a bit can exist in multiple compute states simultaneously, known as qubits.

“What that means is you can test large combinations of ones and zeros across qubits simultaneously, so it allows you to explore potential solutions to very large problems,” Welser says.

One of the key application areas for quantum computing is in chemistry, where Welser says it will have applications in materials science.

“Maybe you’re trying to discover new molecules for pharmaceuticals,” he says. “So let’s take the molecule caffeine, which has 95 electrons on it. If you wanted to do a simulation of that molecule on a standard classical computing system you need to have a supercomputer with 10⁴⁸ bits.

“To give you a reference point, there are 10⁵⁰ atoms in the entire Earth. There’s just no way you could ever build a system big enough to even do that computation.”

However, Welser says with a quantum computer that task could be completed using just 160 qubits. IBM has already released a system with 65 qubits and has plans to release a 1000-qubit system by 2023.

However, in addition to needing to build scale, he cautions that today’s quantum computers are still relatively error prone. Hence IBM is working to improve the error rate and to encode error correction into their quantum systems.

“When we reach that point — 1000 qubits — assuming we have enough error correction control, that should be enough to start doing some interesting things in chemistry and material science,” Welser says.

IBM is already fielding interest in quantum computers from companies including Daimler AG, the parent company of Mercedes-Benz, to model the materials used in next generation batteries for electric vehicles. IBM is also working with the bank JP Morgan Chase to solve optimisation problems in the financial services sector. 

“The companies that are interested are ones that are bumping up against the limits of what they can do with current classical systems, so they want to actually start experimenting with quantum,” Welser says.

While many of the benefits of quantum computing are reliant on the creation of suitably powerful systems, Welser says IBM is also investing in creating a community of people with the skills needed to make use of them. Since 2016, when IBM put the first quantum computer on the cloud, anyone can access the company’s public quantum systems through the IBM Quantum Experience. And organizations can access their premium systems by joining the IBM Q Network. IBM has also been contributing to the creation of an open source software kit called Qiskit which provides a quantum-oriented software programming tools.

“For quantum computing to take off it requires a very large ecosystem of people utilising it, finding how to apply it, learning how to use it, and building better software for it,” Welser says.

Ultimately, he says the contribution that quantum computing can make in fields such as material science will play a vital role in solving some of the most complex challenges facing humanity.

“If you think about a lot of the problems we face today, with the climate, with sustainability, or even fighting pandemics, a lot of it does involve the finding of new materials and new substances that can help us,” Welser says. “If we have systems that can not only do high performance computing like we have today, but also do very large problems and even quantum simulations, all of these combined could really offer a powerful way of discovering new materials.”