Madrid – A quantum computer from Google completed in 200 seconds a task that the fastest conventional computer in the world would take about 10,000 years to complete, according to an article published today in the scientific journal Nature.
This demonstration of quantum supremacy is a milestone for quantum computing, a computer system that aspires to perform tasks exponentially faster than conventional classic computers and with a low error rate.
Unlike classical computers, quantum computers use a system of cubits (quantum bits) that store information in the two digits of the binary code, 1 and 0, while classic machines use bits and must choose between storing data in one of the two figures, 1 or 0.RELATED
With this new protocol, known as overlay, quantum machines can quickly solve very complex problems and process huge amounts of data.
The work, led by Google's quantum computing officer at the University of California, John Martinis, describes the technical steps taken to achieve quantum supremacy.
According to the article, the researchers manufactured a processor composed of 54 cubits, which takes advantage of the superposition and quantum entanglement to explore an computational space exponentially larger than that provided by the classic bits.
A ulna did not work properly, so the device worked with 53 cubits.
The team developed error correction processes to maintain high operational fidelity (up to 99.99%).
To test the system, they designed a sampling task of random numbers produced by quantum circuit. The processor collected a million samples in approximately 200 seconds, which would have led to a next-generation supercomputer about 10,000 years.
"This demonstration of quantum supremacy over the main current classical algorithms in the world's leading supercomputers is really a remarkable achievement," said William Oliver, a researcher at the Department of Computer and Physical Engineering at the Massachusetts Institute of Technology (MIT).
Oliver writes in the same issue of Nature an article published "Quantum computing takes off" in which he also points out that more work must be done before quantum computers become a practical reality.