| Cluster Computing Gets Closer |
|
11 March 2005, for ScienceNOW A radically new path to quantum computing, first proposed 4 years ago, just might work. In the 10 March Nature, physicists report creating a so-called "cluster state" of four quantum-mechanically linked photons--the basis for a new approach called cluster quantum computing. Quantum computers are based on quantum bits, or qubits: individual photons or atoms that reside in a state of superposition, in which they are both "on and off" at the same time. Qubits can be entangled, so that a measurement of one instantaneously affects others. These properties allow quantum computers to run several operations in parallel, promising huge gains in speed over traditional number crunching. There are various ways of building quantum computers, such as using atomic spin. But no one is sure which type of setup might be easiest—or even possible—to scale up to make a useful computer.A new approach is the cluster state quantum computer, which relies on an initial state of many highly entangled qubits. Performing certain measurements on the cluster should change how the cluster processes information, while other measurements ought to make it spit out an answer. Even with just four such entangled qubits, the theory goes, a cluster quantum computer could implement all the basic logical operations. Multiple qubits could be networked to build a more complex computer. Now, Anton Zeilinger of the University of Vienna in Austria and his colleagues have achieved such a cluster state with four entangled photons. Their setup did not allow actual computations because of quantum indeterminacy, in which the same input produced random outputs. But by measuring hundreds of these random outputs, they showed that the photons initially had the entanglement and superposition necessary for cluster computing. The next step is to try storing the entangled photons in fiber optic cables, so that one photon can be measured, and then returned to the cluster nanoseconds later. This should quell the randomness so that given a certain input, the cluster will produce the right answer. Physicist Seth Lloyd at the Massachusetts Institute of Technology calls the experiment "a striking demonstration of the power of entanglement." The study shows that "cluster state quantum computing is not just some theoretical idea that might be virtually impossible to implement," he says. |