CPM Seminar

How to build a quantum computer that might actually work

Austin Fowler

Institute for Quantum Computing
University of Waterloo

A quantum computer processes data using quantum mechanics, enabling it to solve certain problems exponentially faster than is possible using any existing computer. On paper, at least. Actually building a quantum computer is an extraordinary challenge, primarily because errors are common in fragile quantum systems and correcting these errors is a complicated process. Nevertheless, with existing computers now running into fundamental physical limitations that prevent significant further improvement in areas such as their clock rate, the quantum computing community has powerful motivation to overcome their extraordinary challenge.

We give an introduction to one potential hardware realization of a quantum computer, based on loops of superconducting metal, and describe some common error sources. We then describe the standard techniques of correcting quantum errors, and the many restrictions on large-scale computer design that result from using the standard techniques. After presenting a complete computer design within these restrictions, we then show that the maximum tolerable error rates of the individual components that make up the computer may be unachievable low, despite best efforts to tolerate as many errors as possible. This is not the end of the story, it simply gives us motivation to look at a more complicated error correction technique — topological error correction.

We continue with an introduction to topological error correction assuming no prior exposure to the field. We outline why this approach to error correction permits much higher error rates to be tolerated — error rates that are arguably within experimental reach. This leads us to a complete redesign of the computer, which we attempt to give a clear picture of.