The story appears on

Page A8-9

November 19, 2018

GET this page in PDF

Free for subscribers

View shopping cart

Related News

Home » Feature

The race is on for a quantum computer

Huawei recently announced a new cloud service platform that includes a quantum computing simulator, available to the public to allow quantum research and education in the field. Earlier this year, Alibaba Cloud launched its quantum computing service in association with the Chinese Academy of Sciences.

Chinese companies are moving quickly though they are late entries in the quantum mechanics race, as international players such as IBM, Microsoft, Intel and a dozen others have been competing to build the first real quantum computer.

Classic computers store 1 or 0 in a “bit,” while a quantum computer uses qubit (quantum bit) to store a combination of 1s and 0s. One qubit holds two values, two qubits four, three qubits eight and so on. Thus its computing power is exponentially greater than that of “bits.”

The companies are betting big on that estimated computing power, which is said to far exceed super computers and can solve problems in many fields that are impossible today. Though a quantum computer big enough to calculate faster than current super computers has not yet been built, companies and innovators have already taken the hypothesis as granted and are working on possible applications, such as quantum GPS or quantum solutions to overcrowded subway stations.

Shanghai Daily asked two Nobel laureates in physics about their take on this hypothesis and its possible applications. The 2012 laureate Serge Haroche, whose research laid the foundation for quantum computing, is highly suspicious while 2004 laureate and string theorist David Gross waits for it to tackle unsolved problems in atomic physics and many other fields. The hottest topic of all is the combination of artificial intelligence and quantum computation.

“Many problems in physics and other fields depend on the speed of calculation,” Gross said.

“By now, there is no question at some point there will be some kind of quantum computer, now we have them in 50 qubits, but it needs to be hundreds and thousands of qubits to have a powerful one, which is substantially more difficult to be built,” he added.

He estimated that it could become reality in 10 to 30 years, a lot shorter than the previous common expectations of 50 years. As a fundamental theorist physicist, he also explained how a quantum computer can simulate and create a model of the world — which is made of particles — better than classic computers.

“Nothing in the law of physics says it is impossible,” he said. “It is just very difficult.”

One of the difficulties lies in scale. Many big players have built a kind of quantum computer, but quantum physicist Haroche calls those “toys,” small systems that demonstrate the possibility of what it can do when scaled up.

Not easy to scale up

“The good question is not when (a real quantum computer will be built) but if, because I’m not sure it will be,” Haroche said. “A real quantum computer has to be useful and achieve where classic computers fail. To do that, you have to build it in a much larger scale than what has been done up to now.”

But it isn’t easy to scale up.

“You lose the quantum coherence very quickly when you increase the size of the computer,” Haroche said. “Something is missing in the basic science.”

Many companies have been trying to do quantum simulations on classic computers. Alibaba recently released a paper that indicates it might be more competitive against quantum chips, while other companies are comparing their latest quantum chips against regular supercomputers.

“Quantum communication is closer to becoming useful for practical application, maybe in the next decade even, but the question here is why,” Haroche said, referring to the double quantum-teleportation work done by the team led by Chinese scientist Pan Jianwei, who was named by Nature as one of the 10 people who mattered in 2017.

Pan’s vision is to build a global quantum communication network which will utilize satellites as middlemen. And that raises the question of post-quantum encryption, because the current public key system — meaning different codes for encryption and decryption — is certainly vulnerable to a quantum computer.

“We will know in 20 to 30 years whether it is serious,” Martin Hellman, co-inventor of the public key cryptography which won the ACM Turing Award in 2015, told Shanghai Daily.

“But we need to start getting prepared now, because items like our medical records that are encrypted now might be broken in the future.”




 

Copyright © 1999- Shanghai Daily. All rights reserved.Preferably viewed with Internet Explorer 8 or newer browsers.

沪公网安备 31010602000204号

Email this to your friend