A Canadian startup has received a high-profile boost in its goal to become the go-to service for quantum computer builders seeking to measure and improve the performance of their technology.
On Wednesday, tech giant Google LLC announced that it would be integrating software created by Quantum Benchmark of Waterloo, Ont., into its new open-source platform for quantum computing. Developers will be able to use the software as a yardstick for checking the reliability of their results, a pressing issue in the nascent field because of the unstable nature of quantum systems.
“Getting that nod of approval from Google means that our tools are being recognized as important for the assessment of their hardware,” said Joseph Emerson, Quantum Benchmark’s CEO and a physicist at the University of Waterloo. “We’re excited to be working with them.”
The startup is a by-product of Waterloo’s blossoming quantum tech scene, but its initial $1-million financing has been led by Vancouver-based Vanedge Capital Partners Ltd.
Together with a small team of scientists and engineers Dr. Emerson has sought to carve out a vital niche in what is expected to become a major industry built on the unique advantages of quantum computer systems.
Those advantages stem from the nature of matter and light, whose physical properties at atomic scales, such as position and energy, come with a built-in uncertainties. In a quantum computer, the elements of the system that perform a calculation, known as “qubits,” are made up of microscopic entities that are placed in similarly ambiguous states. Instead of representing a 1 or a 0, like a conventional digital computer bit, a qubit can be a little bit of both. This affords a degree of flexibility that gives quantum computers the ability to take shortcuts and vastly speed up the rate at which certain kinds of calculations can be performed.
Potential applications of the technology range from encryption and searches of big data to drug discovery, materials research and machine learning.
The problem is that maintaining a computer in a quantum state long enough to perform useful calculations is notoriously difficult and gets harder as the number of qubits grows. A way to quantify this is to measure the growing rate at which a quantum system tends to make errors as more qubits are added.
To combat this, quantum computers typically employ many times more qubits for error correction than for the original calculation. With the field still in the early stages of development, the need has emerged for tools that can characterize a system’s error rate and help hardware designers optimize their technology.
This is essentially what Quantum Benchmark has brought to the table said Alan Ho, project manager of Quantum AI, a Google research team based in Los Angeles.
“There is a lot of research in this field and Quantum Benchmark is the first company to have industrialized this technology,” he said.
Google is among the players currently working on quantum computing hardware and in March the company unveiled a 72-qubit chip. While still far from superseding a conventional computer, the development is regarded by the industry as a milestone on the road toward a practical quantum computer.
Other players include IBM, which already has a 20-qubit quantum computer that software developers can access on the cloud. Google is moving in a similar direction, and has made its hardware available to a limited number of partners with plans to expand access in the future, said Mr. Ho.
As a precursor, Google has released a quantum framework called Cirq for developers to work with. The Quantum Benchmark toolkit is being made available as a part of Cirq, but the relationship is not exclusive.
“Our endgame is to provide these tools for all quantum platforms,” Dr. Emerson said.
One notable exception is the quantum computer developed by Canada-based D-Wave Systems Inc., which is a different sort of beast that essentially performs its calculations by allowing a quantum state to settle into a stable configuration. The product Quantum Benchmark is putting on the market is aimed at the more universal form of quantum computer that can hop between fleeting quantum states and whose emergence at a commercial scale remains a long-term objective of the industry.
In that context, the product has the potential to make a large impact, said Daniel Lidar, an expert in quantum error correction at the University of Southern California who has studied D-Wave’s machines but who is not affiliated with Quantum Benchmark.
The Waterloo company’s focus “is unique in the quantum-computing startup sphere, and it is encouraging that it is being adopted by quantum hardware developers,” Dr. Lidar added.