First published in AMS business magazine. Author: Douglas Heingartner
Quantum computing marks a new chapter in computer history, as it allows almost infinitely more calculations to be performed than with the digital computers we have today. But so far, most of the investments in the field have gone into quantum hardware, leaving great opportunities in developing the software that will make these powerful machines useful. Software is the main research focus of QuSoft – a joint initiative of the Vrije Universiteit Amsterdam (VU), the University of Amsterdam (UvA) and the Center for Mathematics and Computer Science (CWI), which opened in the Amsterdam Science Park in December 2016. The research centre will collaborate closely with QuTech, an allied research institute in Delft that focuses on the quantum hardware itself.
Faster, more powerful computing
QuSoft is the brainchild of Harry Buhrman, a group leader at CWI and professor of computer science at the UvA. He likens the current situation to the 1960s, when there was much new computer hardware was developed, but no one really knew what to do with it. “People had no clue what we could do with the computer,” says Buhrman, who estimates that powerful quantum computers will start becoming available in five to fifteen years’ time. “You need the hardware, but the software enables us to use it. So it’s just as important, if not even more so.” Quantum computing is based on the idea from physics that the smallest particles, for example electrons, can be in two different states at the same time. So, instead of a bit being either a 1 or a 0, the so-called qubits that form the basic unit of quantum computing are both a 1 and a 0 at the same time (which is also known as superposition). QuSoft can currently work with about nine or 10 stable qubits, which is impressive from a scientific perspective, though not so much in terms of computing power. But Buhrman’s prognosis is that it will “maybe be another five years” before 10 qubits will have grown to 50 qubits, and it’s then that things become more interesting. “With 50 qubits, you can have 250 possible superpositions at once, and that is not so easy to simulate,” he says. “In that instance, you really do need supercomputers, and maybe those won’t even be good enough. Certainly with 70 qubits, we go through the roof of what we can do now. And herein lies the power of quantum computing: massive computing potential, with more calculations than you ever dreamt of.” (As soon as we can master that new, counterintuitive quantum software, that is.)
Safeguarding the future
Another focus at QuSoft is developing new forms of cryptography. Much of today’s cryptography – for example the type that underlies secure online transactions – will, according to Buhrman, “be easy to crack when the first working quantum computers arrive.” He goes on to explain, “Quantum computing will destroy quite a lot of the encryption that is vital for our society to run. Not only for buying goods online, but also for military applications or diplomatic communications. A lot of this cryptography is broken when you have a quantum computer.” So it’s important, says Buhrman, to start developing cryptographic applications now that will be able to withstand those new computers, which will be able to retroactively decrypt messages that were secured using the old methods.
When deciding upon its location, it made perfect sense to choose Amsterdam as the base for QuSoft. Quantum computing requires collaboration between various disciplines, something that Amsterdam excels at. The city also has a long history of quantum research. “We started working on quantum computing very early on. It’s not so easy to get into; you need to know about computer science, physics and quantum mechanics. And there aren’t a lot of people who are trained to do all those things. But since we, in Amsterdam, have worked on this topic for many years, we do have a lot of highly trained people.” The new centre builds on the excellent reputation of the participating Amsterdam institutions, and also strengthens the position of the Netherlands as a centre of world-class quantum computing. According to Buhrman, “QuSoft is in one sense very academic, because we carry out academic research to develop quantum algorithms,” but there are also obvious benefits for industry and corporations. Though it’s not yet exactly clear what those will be, Buhrman does say, “everyone somehow feels intuitively that the benefits will be great. If what we do can eventually lead to something very useful and applicable, then that’s terrific.”