IBM wants to build a universal quantum computer for commercial use
IBM has announced plans to build the world’s first “universal” quantum computer this year. Dubbed IBM Q, the system will be accessible to select early access industry partners over the internet for a fee.
This universal quantum computer would also be the first of its kind to be used commercially, and IBM thinks that it could play a key role in developing a market for future quantum machines that can perform complex calculations beyond existing classical computers, particularly in the fields of chemistry, machine learning, and finance.
As an example, IBM cited its own Watson supercomputer, where the latter would process existing data on drugs to help researchers create better ones, and a quantum computer would then map the interaction of molecules to help formulate entirely new drugs.
The new project builds on existing knowledge gleaned from IBM’s current Quantum Experience cloud computing service, which is based on an actual 5-qubit quantum computer. Anyone can access the latter for free, and it has enabled researchers to practise building quantum algorithms without needing to have their own quantum computer.
The company foresees its IBM Q quantum computers moving to 50 qubits in the next few years, and a universal quantum system of that scale may be able to perform complex calculations at a rate that even today’s top multi-Petaflop supercomputers cannot emulate.
IBM isn’t alone in this endeavor, and the six founding members of the IBM Research Frontiers Institute, comprising Samsung, JSR, Honda, Hitachi Metals, Canon, and Nagase, are also exploring quantum computing applications for their industries.
IBM’s end goal is to put a new realm of computational power within the reach of businesses and organizations, thus enabling them to solve real-world problems. But the most exciting is that no one actually knows the full extent of the capabilities of a universal quantum computer, and we could be in for some very unexpected applications and discoveries.
Source: IBM Think Blog