Tokyo-based NTT Gives Free Access to a New Prototype Quantum Computer in the Hope Engineers Will Help Improve It
Yesterday Nippon Telegraph and Telephone (NTT) shared its prototype quantum computer for public use over the internet. A joint, state-sponsored research project with Japan’s National Institute of Informatics and the University of Tokyo produced the machine. IBM, Google, and others are building prototype universal quantum computers based on superconducting circuits. The Japanese team, on the other hand, created a quantum Ising machine that “exploits the properties of light” to solve combinatorial optimization problems super fast (technical term). Think something more like D-Wave‘s annealer.
How does this magic light machine work? It consists of three main components: a one-kilometer optical fiber loop, an optical amplifier, and an electronic circuit module. A pump sends a light pulse into the fiber loop. The amplifier receives and amplifies the incoming light. The electronic circuit creates a new pulse using an Ising model which is pumped into the loop on top of the original pulse. This process is repeated once each time around the fiber loop until the combined pulse settles into one of two “states” (wave phase of either zero or pi). At that point, the calculation is complete.
All That and a Bag of Chips
This machine can solve a problem roughly 37 times faster than one of Riken’s “small supercomputers” can solve the same problem. The prototype rivaled the speed of D-Wave’s annealers when running certain specific calculations. Besides being lightning fast, the Japanese prototype is capable of running 24/7 at room temperature and only draws about 1 kilowatt of power (similar to a microwave oven). In contrast, superconducting prototypes require temperatures near absolute zero. D-Wave’s 2000Q draws 25 kW and a traditional supercomputer draws around 2500 kW. If proven to remain stable during extended use, this would give Japan’s device a huge advantage. Not bad for a free online prototype.
Starting in January 2018, Japan is setting aside 30 billion yen (around $267 million) for quantum computing over ten years. If proven valuable, this project is sure to get some of that funding. Since the real-world applications for this technology include reducing traffic congestion, optimizing cell phone communications, and discovering new drugs, I’d say its a safe bet. With the right backing, this research could propel Japan to the front of the emerging commercial market for quantum machines.