Quantum computers promise to revolutionize the way and the speed at which programs are done. Applications are finally coming out of the lab, and while a quantum PC is probably still decades away, the latest breakthrough makes that goal significantly closer.
A team from the University of Maryland led by Shantanu Debnath has developed a small but perfectly reprogrammable quantum computer, a first in the field. So far, other quantum systems have been set up to perform just a single function, so this new system adds versatility to power.
The scientists used one of the most reliable quantum architectures available – an ion trap. The computer qubits, the quantum version of the regular one and zero bits, were made by five individual ytterbium ions locked in a straight line by electrodes and a magnetic field. The qubits were then manipulated by lasers, which can entangle two or more of them at the same time. This set up is used to perform the logical operations behind computer programs.
The team has developed a compiler, which looks at the quantum program and then manipulates the five atoms (the hardware in this system) to perform the operation. The newly developed system has an average accuracy of 98 percent, which is great in terms of quantum computing but not so great in terms of regular computing. You really don’t want an error every 50 operations.
The current challenge of quantum computers is not to outperform regular computers (that day will come), but to show that the current approach is viable. Scientists and engineers in the field are trying to understand as much as possible about the effects at play when we manipulate quantum mechanical systems. Harnessing the full power of quantum computing can only come when the footing is solid.
This development is definitely interesting, but the quantum computer of the future needs to be bigger than just five atoms. In the paper, published in Nature, the authors have suggested that it can be scaled up with more atoms and more devices working together.
Scientists are not putting all their quantum eggs in one basket, though. For example, systems that use superconducting circuits can now compete with the trapped ion devices. But we now have to wait and see which technology will be able to bring errors down to the amount we can ignore. We really don’t want a quantum blue screen of death.
1