RSS
Логотип
Баннер в шапке 1
Баннер в шапке 2

Jiuzhang (quantum computer)

Product
Developers: University of Science and Technology of China
Date of the premiere of the system: June 2023
Branches: Electrical and Microelectronics
Technology: Supercomputer

Content

2023

Jiuzhang 3 announcement

On October 11, 2023, researchers from the University of Science and Technology of China in Hefei (an urban district in Anhui province) announced the development of a photonic quantum computer JiuZhang 3. It is claimed that the system is capable of solving ultra-complex mathematical problems in millionths of a second, while the world's most powerful supercomputer will take billions of years.

The project is led by scientist Pan Jianwei, a leading specialist in China's national quantum research program. The first machine of the JiuZhang series was created in 2020. It uses photons as a physical medium for calculations - fundamental particles moving at the speed of light. Each photon carries a qubit - a quantum bit, which is the smallest unit of information in a quantum computer. In the first version of JiuZhang, 76 qubits were involved, in the second - 113. The JiuZhang 3 platform operates with 255 qubits.

Researchers report development of photonic quantum computer JiuZhang 3

The researchers used Jiuzhang 3 to solve a complex problem based on sampling Gaussian bosons, which simulates the behavior of photons passing through a maze of crystals and mirrors. Initially, this task had no practical purpose, but then it was shown that the selection of bosons could find application in cryptography. During the experiment, the Jiuzhang 3 computer solved the problem at the highest difficulty in one millionth of a second. For comparison: the world's fastest supercomputer Frontier with a speed of 1.194 Eflops will take more than 20 billion years to do this.

But despite the colossal performance in solving certain problems, quantum machines as of 2023 are not able to replace conventional computers. In an interview with Newsweek in 2019, quantum computing pioneer Peter Zoller said that despite the emergence of small quantum computers with dozens of qubits in laboratories, a breakthrough in error correction technologies is needed so that such systems are ready for practical use.[1]

Quantum computer announcement

In early June 2023, scientists from the University of Science and Technology of China said they had reached another milestone in quantum computing, saying their Jiuzhang device could perform tasks commonly used in AI 180 million times faster than the world's most powerful supercomputer.

The problems solved by their quantum computer can be applied to data retrieval, biological information, network analysis and chemical modeling, the researchers said. Their paper was published in the peer-reviewed journal Physical Review Letters.

Jiuzhang (quantum computer)

In the experiment, the team used the first photonic quantum computer to solve a problem that is complex for classical PCs. More than 200 thousand samples were used to solve the problem. The researchers used a quantum computer for the first time to implement and accelerate two algorithms -- random search and simulated annealing -- that are widely used in the field of AI.

According to Chinese experts, the fastest classical supercomputer in the world would spend 700 seconds on each sample, that is, it would take almost five years to process the same number of samples. It took Jiuzhang less than a second.

Since the basic information of a quantum computer can represent all possibilities at the same time, in theory they are much faster and more powerful than ordinary computers that users use in everyday life. But the subatomic particles underlying this technology are fragile, short-lived, and prone to error if subjected to even minor environmental disturbance. Most quantum computers operate in extremely cold and isolated conditions to avoid glitches.

Jiuzhang, named after the 2,000-year-old Chinese mathematical text, uses light as a physical medium for computing. Unlike other quantum computers, it does not need to operate hermetically at extremely low temperatures and can run stably longer.[2]

Notes