The Million Qubit Challenge

By Casimer DeCusatis


Within the past few years, quantum computing has moved from the realm of the purely theoretical into the practical as the world’s first working quantum computers have been demonstrated. Currently, these computers are about as powerful as the early difference engines built out of metal gears and hand cranks by Charles Babbage. But if we can harness polarized photons for quantum calculations, there is enormous potential to scale these devices to much more powerful systems. OFC has always been at the leading edge of optics research and development, and this year’s conference (moved to June because of the ongoing pandemic) will host a symposium on quantum information science. In this blog, we’ll take a look at how some researchers are harnessing photonics in an effort to tap into the full potential of quantum computers.  

Why Quantum Computers?

Classical computers represent information as binary numbers (either one or zero) using voltage signals on transistors in electronic circuits. As useful as these computers have become, there are still some types of problems which cannot be solved in less than exponential time on a classical digital computer. Some of these are very interesting, practical problems, such as modeling new types of chemical and biomedical systems or encrypting and decrypting vast amounts of data.  For these problems, quantum computers offer a tremendous advantage, namely the ability to complete calculations in quadratic time (billions of times faster than a classical computer). This requires representing information not as ones and zeros, but as probability amplitudes; just as a bit is the fundamental unit of computation in a digital computer, a qubit is the building block of a quantum computer.

How Many Qubits Are Required

Generally speaking, the more qubits we can process in a single computer, the more powerful our computations become. Unfortunately, qubits rely on quantum interference effects and are very unstable. It’s a huge technical challenge to make qubits, then hold enough of them stable for long enough to run an algorithm before they spontaneously break down. The largest quantum computes in existence today have less than 100 qubits. While this is enough to perform some impressive feats (Google recently achieved quantum supremacy on a machine with 54 qubits), unlocking the true potential of quantum computing would require a much larger system.

Quantum Computer Architecture

There are many ways we can physically realize qubits, in an effort to build larger, more stable systems. Quantum computers such as the IBM Q System One use superconductors cooled to a few millikelvin (about the temperature of outer space) to manipulate weak electronic and microwave signals. Extremely low temperatures are required to minimize thermal noise that can cause qubits to break down or decohere; elaborate, expensive refrigeration equipment makes up most of the bulk for such devices. Further, quantum computers require extensive shielding from any source of background electromagnetic radiation that could distort qubits used during the calculations. An alternative quantum computer architecture is the trapped ion design used by companies like Honeywell and IonQ. Precision lasers remove outer electrons from atoms of a rare-earth isotope (such as ytterbium) to create an ion, which can be manipulated by other lasers and oscillating voltage fields. This type of qubit should be able to maintain its quantum state for a relatively long time compared with superconducting qubits. A third option under investigation is the use of polarized photons to represent qubits, resulting in a system which can operate at room temperature and keep qubits stable for even longer periods of time.  

Photonic Quantum Systems

The long-term stability of photonic quantum systems is a significant advantage compared with other approaches. A photon can be polarized horizontally or vertically to represent a one or zero, as well as diagonally to represent a superposition of these two states. The resulting polarization states are very stable; for example, we can detect the polarization of light from distant stars which travels for billions of years through space, essentially unchanged. Of course, photons also have their drawbacks as a computational medium. Unlike superconducting qubits or ion traps that remain stationary, photons by definition are always in motion at the speed of light. Some researchers have called photonic quantum computers analogous to manipulating something as small as a virus as it speeds past you at 300,000,000 meters per second. 

Researchers are motivated to grapple with this problem, because a photonic quantum computer with a million qubits might be possible within a decade or so, and perhaps even sooner. At least one startup, with help from investors like Microsoft, has so far raised over $215 million venture capital towards the development of a million qubit photonic quantum computer. Such a device could allow us to simulate physical systems well beyond the capacity of any digital supercomputer in the world, or to disrupt the public-private key encryption which forms the basis of a trillion-dollar global online economy.  We can only speculate on the type of problems that such a device could solve, and photonics might make it possible to build the first machine of this kind sooner than we ever thought possible.  The use of photonics in quantum computer systems is understandably a much anticipated topic for the next OFC conference, and you won’t want to miss out on the latest developments in this field.

What would you do with a million qubit quantum computer?  Drop me a line on twitter (@Dr_Casimer), and maybe we’ll use your ideas in a future blog.

 

Check out the content of the symposium, Quantum Information Science and Technology (QIST) in the Context of Optical Communications​, to be held at OFC 2021.

 

 

Posted: 11 January 2021 by Casimer DeCusatis | with 0 comments

Comments
Blog post currently doesn't have any comments.
 Security code


The views expressed in this blog are those of the authors and do not necessarily reflect the views or policies of The Optical Fiber Communication Conference and Exposition (OFC)  or its sponsors.