When fully realized, quantum computers will power humanity through problems that seem impossible to solve today. But even impossible problems have a way of proving possible, given enough time: much like humans being able to fly seemed impossible before the Wright brothers rewrote the future.
One of the leading limitations in today's qubits is that they can very rapidly decohere - going from a state where they provide practical work towards one where the calculations don't provide accurate results. So it's another type of race against time, one where researchers with the Japanese Institute for Molecular Science have now leaped to first place by shattering the previous record for the fastest two-qubit gate operation ever done in quantum computing.
Qubits, as the name implies, are the quantum equivalent of the binary bit that's powered our technological revolution. The particular power of qubits is that they don't need to be fixed at a value of one or zero. Instead, they have the added ability to be able to represent both one and zero. It enables qubits to provide much more work per unit of time than the basic bit. It has already allowed real-world computations (such as BMW's Sensor Placement Challenge) to do in six minutes what would take our most powerful computers exponentially longer.
A two-qubit gate operation is the most fundamental (and first in the scale of advantageous) qubit arrangement, and it requires that the two qubits be entangled - simplifying things immensely; this essentially means that their state is a shared (or coherent). As we've seen, however, today's quantum systems are prone to noise (such as environmental radiation, among others). Noise can lead their entanglement to decohere, which will fumble whatever operation they're running (remember when you overclocked your PC too much, and Prime95 returned an error? That's one way of putting it.
There are two ways to deal with this issue: we either perform the operations faster - before decoherence has time to set in, generally at the microsecond scale - or increase the qubit's entanglement longevity. The Japanese researchers went with the former approach.
Using lasers, the researchers cooled two atom-qubits made from the element Rubidium (being the absolute smallest particles of a fundamental unit, atoms are naturally inclined to quantum duties) to temperatures near absolute zero (−273.15 °C).
It isn't the only absolute-zero technique for handling qubits; the physics of it has to do with how fast the molecules interact with each other. At higher temperatures, they interact faster and are more excitable. Cooling them to the equivalent vacuum of space, on the other hand, is akin to putting them into hibernation, slowing down their interactions with each other and the environment itself, thus increasing coherence times. Of course, like a bear, they'll still jump out of that state with a big enough shove, but maybe they can handle a pinprick or two.
The researchers then secured these atoms within a micrometer of each other using optical tweezers, and a final laser manipulated the qubits at ten picoseconds (one trillionth of a second) intervals. Using this technique, the researchers successfully ran a quantum gate operation, which concluded in 6.5 nanoseconds - less than half the previous-fastest two-qubit gate operation, which took 15 nanoseconds.
A thousand nanoseconds fit within a single microsecond, so there was plenty of time between the qubits being entangled and the system decohering to perform calculations.
While the researchers' work doesn't solve the problems with quantum computing just yet, it's a step in the right direction. At the very least, it shows that there are still faster operating speeds to be unlocked in the realm of quantum, which should ultimately scale the available performance from this new, emerging computing solution.
There are certain practical caveats to the system created by the researchers. For one, they only managed to entangle and operate on two entangled qubits. IBM, for instance, plans on introducing its 433-qubit Osprey Quantum Processing Unit (QPU) this year.
Another thing to note is that the rubidium-atom qubits employed by the researchers - and the technique that allowed for breaking the world record - require cooling the system towards absolute zero. That's a costly endeavor and a hard one to replicate in High-Performance Computing (HPC) and other environments across the world.
There are many runners, and certain tech will undoubtedly be developed slower than others, leaving it to eat the proverbial dust of capital and time investment. But until there's a qubit technology that's the clear leader - much like silicon was for semiconductors at the time of their introduction - the question shall remain open-ended.
