Honeywell Quantum Solutions has been relatively secretive about what it was working on since it entered the world of quantum computing with modest fanfare in mid-2018. Their chosen technology was trapped-ion quantum computing.
At the time, Tony Uttley, President of Honeywell Quantum Systems, quietly predicted it would eventually become a leader in quantum computing. His confidence was based on Honeywell's legacy and expertise in most of the technologies required to build a trapped-ion quantum computer. Serving a wide range of industries, Honeywell’s various divisions have built state-of-the-art hardware and software control systems, advanced electronics, optics and photonics, which mainly encompasses lasers, modulators, fiber optics, ultra-high vacuum environments and cryogenics.
With its assembled team of atomic, molecular and optical physicists, scientists, engineers, and technicians, Honeywell Quantum Solutions expects to claim the title as the world's most powerful computer in the next three months as a result of a quantum computing performance measurement developed by IBM in 2017 called Quantum Volume.
Quantum volume is just one of the major contributions IBM has made to the entire quantum computing ecosystem. It was developed as a research tool for the entire circuit-based quantum computer community to assist in the optimization of various configurations and fine tuning of their quantum computers.
Quantum volume is a numerical value that indicates the relative complexity of a problem that can be solved by a quantum computer. The number of a quantum computer’s calculating bits, called qubits, and the number of operations that can be performed are called the width and depth of a quantum circuit. The deeper the quantum circuit, the more complex of an algorithm the computer can run.
Quantum Volume analyzes the collective performance and efficiency of things like the number of qubits, how qubits are interconnected, gate and measurement errors, circuit compiler efficiency, and more, then produces a single, easy-to-understand number. The larger the quantum volume number, the more powerful the quantum computer.
Is quantum volume the right measurement to compare quantum computers? Yes, if anything, Honeywell’s quantum volume might be somewhat understated by the metric because its trapped ion qubits have a greater coherence than superconducting qubits and can allow a greater circuit depth than allowed for in the calculation of quantum volume.
How powerful is Honeywell’s quantum computer?
To put Honeywell’s claim in perspective, we need to look at IBM which has a stated goal of doubling its systems' quantum volume every year. Although IBM uses superconducting qubits versus Honeywell’s trapped-ion qubits, the quantum volume comparison is comparable.
IBM has met its doubling goal every year since its first quantum volume measurement of 4 was made in 2017. This year IBM upped its quantum volume from 16 in 2019 to a quantum volume of 32 in January 2020.
As of the announcement date, Honeywell Quantum Systems quantum computer has a quantum volume of 16. In three months, Tony Uttley stated Honeywell's quantum machine will have a quantum volume of 64. At that time, according to quantum volume, and unless another company announces a higher value, Honeywell will have the most powerful quantum computer in the world.
To put that number in perspective, I believe that IBM probably already knows how to achieve a quantum volume of 64, likely much higher. However, Honeywell’s next claim is probably out of reach for every other circuit-based quantum computer.
Honeywell’s Quantum Volume is predicted to grow by an order of magnitude every year
Uttley explained that Honeywell’s goal is to increase the quantum volume by one order of magnitude every year. That means in 2021, Honeywell's trapped-ion quantum computer will have a quantum volume of 640. Then by 2022 it will have a quantum volume of 6400.
In a previous article, I estimated quantum computers would have a quantum volume of 1000 by 2025, which would allow them to solve many problems that classical computers would require an inordinate amount of time to solve or couldn’t solve at all.
Uttley labels this stage of quantum computing as “classically impractical.” It is also known as Quantum Advantage. I asked Uttley when he thought that would occur. Although he declined to give me a year, he provided other guidance.
“I think it's going to be a lot sooner than we think, “he said. “And I think that for two reasons. One is, we're going to be expanding the capability and adding a lot of quantum volume fast which is going to allow us to go after important use cases. And the second is the beauty of quantum computing - it's exponential.”
A Black Swan with a positive impact
No one was expecting a quantum computer with a quantum volume of over 600 to appear on the horizon this soon. It will take time for algorithms and other parts of the quantum ecosystem to catch up.
Scott Aaronson is the David J. Bruton Centennial Professor of Computer Science at The University of Texas at Austin, and director of its Quantum Information Center. I asked him what he thought the impact of a system with a quantum volume of about 600 would have on the quantum industry and business. His reply was interesting and confirms some catchup will be required to fully utilize a large quantum volume:
"We don't know whether there are any business applications for the sort of near-term quantum computing that you're talking about. There are some things that are worth trying, like simulations of quantum materials or generating certified random bits. But even if they turn out to yield a quantum advantage, the path from there to actual business application is not direct."
Advantages of starting with trapped ion
The fact that Honeywell Quantum Solutions started with trapped-ion qubits gave it several advantages over other qubit technologies:
- Trapped ions can maintain their quantum states much longer than superconducting qubits
- Both single and two-qubit gates have very high fidelities
- Qubits are fully connected
- High-resolution single bit rotations
- Mid-circuit measurement and reset for conditional if statements (unique to Honeywell Quantum Solutions)
- There are no physics-based limits to scaling to high qubit counts
- Trapped ion has a path to error correction
- Natural qubits are all identical
There are also a few disadvantages to trapped-ion qubits. One, scaling, seems not to be an issue when looking at Honeywell’s future technology plans. The other is slower gate speeds.
Honeywell Quantum Solutions’ roadmap to the future
Since Honeywell’s quantum computer is not limited by either error or connectivity, moving from a quantum volume of 16 to 64 is a simple matter of adding qubits - something that the ion-trap architecture can accommodate.
Next, moving up from a quantum volume of 64 to an order of magnitude higher quantum volume of 640 in a year is possible by adding additional qubits while retaining full connectivity of the qubits and maintaining the low error rates they have already achieved.
Incredibly, the next several years of order-of-magnitude improvements are designed to be implemented in the same trap that Honeywell is initially releasing. While this step involves improving error rates, Honeywell Quantum System researchers have already proven that achievable in dedicated test systems.
Looking a few more years out in the future, Honeywell Quantum Solutions researchers are already testing advanced 2-D architectures, expanding on their implementation of the QCCD architecture. A preprint research paper on this technology was recently published.
Illustration of Honeywell’s programmable QCCD quantum computer system with a photograph of the trap.(b) shows the load hole (black), load zone (purple), storage zones (orange), gate zones (blue), as well as auxiliary zones (yellow) for additional qubit storage. Source: Honeywell Quantum Solutions
And much further out, to ensure continued support of the company’s long-range goal of increasing quantum volume by one order of magnitude each year, Honeywell Quantum Solutions is working on integrated technology for ion control, such as integrated electronics, light detectors, and optics.
Tony Uttley nicely sums up how and what Honeywell Quantum Solutions has brilliantly accomplished in such a short time and where its technology is headed in the future:“The trapped ion technology that we're using starts with perfect qubits. It is reinforced by the fact that we control our own architecture by building our own traps, by developing our own control system, and by our ability to do such precision control. Quantum CCD architecture has been talked about for trapped ions for decades, but no one's been able to do it. That's what we are doing, that is what we have done to get to where we are now, which is why we are so confident that the same trap that allows us to achieve a quantum volume of 64 is the exact same trap that will allow us to do a quantum volume of 640,000.”