Quantum computing takes a step toward commercial viability
Eric Olson | January 30, 2019IBM revealed its newest quantum computer at the 2019 Consumer Electronics Show in Las Vegas. The company calls the IBM Q System One “the world’s first integrated universal approximate quantum computing system designed for scientific and commercial use.”
That might be a bit of over-embellished marketing speak. Q System One is not a true universal quantum computer – if "universal" means a machine designed not only for solving specific problems but one with the ability to simulate any real, physical quantum system as well as having all of the computational capabilities of a classical computer.
But Q System One is an important step forward in quantum computing that advances the field's commercial viability. And the machine is capable of more general-purpose computations than competing designs like D-Wave’s quantum annealing computers.
System details
Q System One is a fusion of cutting-edge technology, industrial design and systems engineering created by IBM scientists in collaboration with three design studios: Map Project Office, Universal Design Studio and Goppion. The device consists of a multi-level cylindrical dilution refrigerator filled with a neatly arranged array of looped coaxial lines, signal amplifiers and cryogenic isolators suspended within a nine-foot cube of borosilicate glass.
(Discover RF amplifiers and RF isolators on Engineering360.)
The dilution refrigerator cools one of IBM’s fourth-generation 20-qubit quantum processors to near absolute zero. The design is aimed at minimizing any sources of disruption – including heat, vibrations and electromagnetic radiation – that would interfere with the quantum effects that allow the processor to function.
Q System One advances IBM’s quest to improve the reliability of quantum processors. Although its 20-qubit chip has fewer “quantum bits” than a 50-qubit processor the company tested in November 2017 and less than Google’s 72-qubit chip, raw qubit numbers do not necessarily lead to higher processing power. Even more important at this early stage of quantum computing is reducing the error rates of quantum processors. Improving quality is one of the keys to achieving what IBM refers to as “quantum advantage” and what Google calls “quantum supremacy” – quantum computers that solve problems faster and more efficiently than traditional computers.
Superposition and entanglement
Although that feat has yet to be demonstrated, quantum computing holds the promise of incredible computing power. Quantum computing’s potential arises from the strange, unintuitive behavior that occurs in the small-scale world of atoms and subatomic particles, including the quantum mechanical phenomena of superposition and entanglement.
Superposition allows qubits to exist in multiple states at once, while entanglement – an effect Einstein called “spooky action at a distance” – couples qubits together so that there is a correlation between their behaviors. Unlike classical binary bits which store information as distinct ones or zeros, qubits hold data as a superposition of ones and zeros.
In theory, an n-qubit quantum processor can exist in a superposition of 2n states so that a pair of qubits can evaluate four possibilities simultaneously (22 = 2 x 2 = 4), three qubits can evaluate eight possibilities at once (23 = 2 x 2 x 2 = 8), and so forth. A quantum computer with 300 qubits could theoretically exist in a superposition of more states than there are atoms in the universe.
Quantum advantage
Scientists hope to harness this power to solve a variety of computationally intensive real-world problems that are beyond the capabilities of traditional computers. With functional quantum computers, researchers may be able to discover new medicines by simulating the complex interactions that take place between proteins; uncover novel materials by modeling molecular chemical reactions; supercharge artificial intelligence techniques like neural networks to efficiently process massive data sets; identify risk factors in the global economy and pinpoint ideal investments; and improve the global supply chain by optimizing shipping paths and schedules.
These issues have been identified as areas where quantum computing may enable rapid progress. Success, however, is not a foregone conclusion. To realize quantum advantage, scientists must develop increasingly accurate quantum processors with lower error rates and more qubits.
Equally important, scientists must find a way to model and program the challenges at hand in a way that exploits the quantum effects that make quantum computers so powerful. Standard programming techniques used to manage the binary bits of classical computers must be altered to harness the qubits of quantum computers. Such quantum algorithms will provide a quantum processor with instructions to encode all 2n superpositional states onto qubits and then simultaneously evaluate all of the possibilities to generate a solution.
A quantum computing network
To give researchers a chance to experiment with and familiarize themselves with the emerging field of quantum programming, IBM introduced the IBM Q Experience in May 2016. In this free, publicly available, cloud-based sandbox, researchers can upload instructions to real quantum computers housed in IBM Quantum Computing Labs around the world and receive the results back over the internet.
The 107-year-old company is not offering its quantum computing resources as a charitable donation to humanity, however. IBM is aiming to capture a share of what could be a very important market by positioning itself as a frontrunner in the emerging field of quantum computing and drawing leading experts in the domain into its ecosystem.
The Q Experience exists alongside the IBM Q Network, of which Q System One will be a part, bolstering its ranks of quantum computers. The Q Network offers businesses and research labs cloud-based access to IBM’s most advanced quantum computers and development resources. It has attracted several well-known companies including JPMorgan Chase, ExxonMobil, Samsung, Barclays, Honda as well as top universities and research centers like MIT and CERN.
IBM Q Hubs, collections of institutions with access to the Q Network pursuing progress in quantum computing, have also been established. One such hub consists of the Oak Ridge, Argonne, and Lawrence Berkeley National Laboratories as well as Fermilab.
The Q Network serves as both an incubator for advances in quantum computing that lead to real-world applications and as a magnet to attract companies to IBM’s cloud-based quantum computing ecosystem. IBM’s aim is to encourage the creation of real quantum advantage and capture a substantial portion of the revenue from what could become a significant sector of the computing industry. The company, however, faces competition from other major technology players. Google is developing its own superconducting quantum processors and quantum algorithms to improve artificial intelligence, while Microsoft is advancing a topological quantum computer and a quantum programming language, Q#.
The competition is driven by the potential of quantum computers to vastly improve computing performance for some of the most challenging and impactful computational problems of our time. Even though quantum computers have not yet demonstrated superiority over classical computers and a truly universal quantum computer has not yet been developed, the march of progress continues. It is, perhaps, only a matter of time before quantum computing fulfills its promise and reigns supreme.
Great, but can it add 2-plus-2 and get the correct answer?