Quantum computing is based around the premise that computers will use individual atoms and molecules in order to process instructions and use memory. Moore's Law states that the number of transistors on a microprocessor doubles every eighteen months, which if we extrapolate forwards, means that somewhere between the year 2020 and 2030, the circuits on a typical microprocessor will be measured on the atomic scale. From here, the next step is to create quantum computers, which have the potential to be significantly quicker at certain types of calculation compared with conventional computers. Given Google's propensity to research and develop emerging technologies, it should be no surprise to discover that Google have a Quantum AI (Artificial Intelligence) team, working towards understanding the physics that will govern quantum annealers. The reason why AI is involved is because artificial intelligence is seen as requiring significant computational power: quantum computing will be able to compute huge numbers of calculations simultaneously, making the technology highly desirable for handling big data as well as artificial intelligence systems.
Google have today released a paper detailing some of their insights into quantum computing and how they have successfully applied proof-of-principle optimization into the D-Wave 2X quantum annealer that Google operates as a joint venture with NASA.Their research showed that to solve a problem containing close to 1,000 binary variables, quantum annealing is materially quicker than conventional computing systems (known as simulated annealing). The technology was found to be more than one hundred million (100,000,000) times faster than simulated annealing running on a single core. The chart at the bottom of this article compares the performance of Simulated Annealing (SA on the chart) with the Quantum Monte Carlo (QMC) and D-Wave 2X technologies. The D-Wave computers are considered to be the closest machines to quantum computers, and as the chart shows the D-Wave technology is much quicker than simulated annealing. Google is confident that this higher performance can be utilized in order to drive research into artificial intelligence systems.
As to how close we are to seeing a proper quantum computer, Google's blog says this on the matter: "I would say building a quantum computer is really, really hard, so first of all, we're just trying to get it to work and not worry about cost or size or whatever." However, there are very real-world applications for this sort of concurrent high performance computing. Big data can mean many things, such as image recognition, operating any sort of traffic control system as used by air traffic control, railroad operators and city planners). We are still several years from the technology being commercially available and are not going to see an Android device running with the technology just yet.