First quantum computer to pack 100 qubits enters crowded race

First quantum computer to pack 100 qubits enters crowded race

The insides of an IBM quantum computer show the tangle of wires used to control and read its qubits.Credit: IBM

IBM’s newest quantum computing chip, unveiled on November 15, has somehow reached a milestone: it contains 127 quantum bits (qubits), making it the first device of its kind to reach 3 digits. But this achievement is just one step in an ambitious program spurred by billions of dollars of investment across the industry.

The “Eagle” chip is a step towards IBM’s goal of creating a 433-qubit quantum processor next year, followed by a 1,121-qubit processor, named Condor, by 2023. Such goals make echoes those the electronics industry has set itself for decades to miniaturize silicon chips, says Jerry Chow, head of IBM’s Experimental Quantum Computing Group at the Thomas J. Watson Research Center in Yorktown Heights, New York.

Other companies, including tech giants Google and Honeywell, as well as a slew of well-funded start-ups, have similarly ambitious plans. Ultimately, they aim to make quantum computers capable of performing certain tasks that are beyond the reach of even the largest supercomputers using classical technology.

“It’s fine to have ambitious goals, but what matters is whether they can execute their plans,” says quantum information theorist John Preskill of the California Institute of Technology in Pasadena.

Quantum Advantage

By harnessing the laws of quantum physics to process binary information, quantum computing circuits such as the Eagle chip can already perform calculations that cannot be easily simulated on conventional supercomputers. Google said it achieved such a ‘quantum advantage’ in 20191, using fabricated qubits, like those from IBM, with superconducting loops. A team from the University of Science and Technology of China (USTC) in Hefei last year reported achieving a quantum advantage using optical qubits2; this year it did the same with superconducting qubits3.

But the tasks assigned to these machines were artificial, warn the researchers. “The current state of the art is that no experiment has yet demonstrated a quantum advantage for practical tasks,” says physicist Chao-Yang Lu, who co-led the USTC effort. Solving real-world problems such as simulating drug molecules or materials using quantum chemistry will require quantum computers to become significantly larger and more powerful.

Quantum engineer Andrew Dzurak of the University of New South Wales in Sydney, Australia, thinks that with 1,000-qubit chips such as IBM’s future Condor, the technology could start to prove itself. “It is hoped that some useful and even commercially interesting problems can be solved using quantum computers in this thousand to million qubit range,” he says. “But to do really paradigm-shifting things, you’re going to need millions of physical qubits.”

Smart Challenges

The Eagle chip has almost twice as many qubits as IBM’s previous flagship quantum chip, the 65-qubit Hummingbird. The increase required the team to address several engineering issues, Chow says. To allow each qubit to interact with several others, the researchers opted for an arrangement in which each is connected to two or three neighbors on a hexagonal grid. And to allow individual control of each qubit without an unmanageable tangle of wires, the team placed wires and other components on multiple stacked levels. Chow says that to solve this “conditioning” problem, the researchers relied on experience with 3D architectures in conventional chips. He adds that it was also crucial to find materials that would perform well at the ultra-low temperatures needed to operate superconducting qubits.

But the processing power of a quantum circuit is not limited to the number of qubits it has. It also depends on their speed of operation and their resistance to errors that could confuse a calculation, due for example to random fluctuations. Chow says there is still room for improvement in all of these respects for superconducting qubits.

Error handling is particularly difficult because the laws of physics prevent quantum computers from using the error correction methods of classical machines, which typically require keeping multiple copies of each bit.

Instead, the researchers aim to build “logical qubits” – in which almost any error can be identified and corrected – from complicated arrangements of many physical qubits. The procedures proposed so far typically require each logical qubit to contain around 1,000 physical qubits, though that ratio depends on the intrinsic fidelity — the resistance to error — of the physical qubits, Dzurak says.

Correction of errors

Some other approaches to building quantum computers hope to benefit from qubits with lower intrinsic error rates. That’s one of the potential benefits of using trapped ions as qubits, like the company IonQ, spun off from research at the University of Maryland at College Park, which last month raised more than 600 millions of dollars when it became the first purely quantum company. computer company to publicly trade on the New York Stock Exchange — a deal that valued the company at nearly $2 billion. Rigetti Computing, a start-up from Berkeley, Calif., also went public this year, with a valuation of $1.5 billion.

IonQ co-founder Christopher Monroe, a physicist at the University of Maryland, and his colleagues last month reported a fault-tolerant logic qubit consisting of just 13 trapped ion qubits4although Dzurak claims that its degree of error correction was “still quite far from what is needed for a useful quantum computer, which needs logical error rates well below one in a million”.

The Google team, meanwhile, achieved similar logic error rates using 21 superconducting qubits.5: again, “an important result”, says Dzurak, but still far from what is needed to solve the error correction problem.

But Chow cautions against overemphasizing getting logical qubits. “We won’t have a situation where we flip a switch and say ‘error correction is on’,” he says. “Improving qubit performance is a bigger story than creating logical qubits and dividing everything by 1,000.”

Signal Boost

IBM and others try to gain a detailed understanding of error-related noise in a circuit and then extract it – much like noise cancellation to improve the signal-to-noise ratio in acoustics.

Beyond Condor-level devices, Chow says, circuit designs are likely to become modular, with multiple chips joined by “quantum interconnects.” It’s not yet clear how best to do this – perhaps with the microwave signals currently used for data input and output to superconducting qubits, or perhaps converting quantum information into light-based signals. . “It’s a whole new area of ​​research,” says Chow.

Many researchers believe that the first real applications of quantum computers are likely to be in relatively specialized areas, such as the simulation of molecules and materials, machine learning, and optimization problems in industries such as finance. To get to this point, “I expect to see a gradual improvement in performance rather than a sudden leap forward,” Preskill says. “It will probably be a long job before we can run useful applications.”