According to a paper published Tuesday in the peer-reviewed journal Physical Review Letters, the JiuZhang 3 prototype improved calculation speed by a factor of 1 million, also breaking records set by its predecessors in the series.
The research team is led by Pan Jianwei, a lead scientist in China’s National Quantum Research Program and from the University of Science and Technology of China in Hefei, eastern Anhui province.
The first Jiuzhang machine, named after an ancient mathematics textbook, was built by Pan’s team in 2020. The series uses photons (tiny particles that travel at the speed of light) as the physical medium of computation, each carrying a qubit. The basic unit of quantum information.
After increasing the number of photons in the first two versions of the machine from 76 to 113 in each, Pan and his team achieved progress to 255 in the latest iteration.
US guru says China’s supercomputing power could surpass all countries
US guru says China’s supercomputing power could surpass all countries
The researchers used Jiuzhang 3 to solve a complex problem based on Gaussian boson sampling that simulates the behavior of light particles passing through a maze of crystals and mirrors.
Although this task was originally proposed as an aimless physical game, some recent research suggests that boson sampling may have applications in cryptography.
In experiments, the researchers claimed that Jiuzhang 3 was able to solve the most complex sample problems and process the tasks within a millionth of a second.
Frontier, the fastest classical supercomputer developed in the United States and named the world’s most powerful in mid-2022, would take more than 20 billion years to complete the same task, researchers said.
China, the United States, and other countries are in a fierce race to achieve “quantum supremacy,” the point at which machines can outperform classical computers and solve problems beyond the capabilities of conventional machines.
But they use different methods to approach their goals, and photonic processors are just one of several types of quantum computing.
Toronto-based company Xanadu also competes with light-based systems. In collaboration with the US National Institute of Standards and Technology (NIST), the company announced in 2022 a quantum processor called “Aurora” with 216 photons.
But despite their speed, these machines aren’t trying to replace common computers. At this stage, you can work on very specific tasks for short periods of time in a protected environment. They also make many mistakes.
Students save quantum computers – but why were they still working at 2am?
Students save quantum computers – but why were they still working at 2am?
When Jiuzhang 2 was launched in 2021, Pan said the team hoped to achieve quantum error correction with “four to five years of effort.”
Quantum computing pioneer Peter Zoller said in a 2019 interview with Newsweek magazine that even though small-scale quantum computers with dozens of qubits have appeared in labs, it is difficult to put them into real use. He said that a breakthrough in error correction is needed to make this possible.
“In the next five to 10 years, we could see quantum computers with hundreds of qubits, some with partial error correction.”
In the same interview, Zoller of the University of Innsbruck estimated that large-scale, fast quantum computers would require tens of thousands or even millions of qubits, along with robust error correction. Did.