“Quantum computers promise to solve problems that conventional computers cannot, because qubits can exist in multiple states at the same time. Using this quantum physics phenomenon, qubits can perform large amounts of calculations simultaneously, which can greatly speed up the speed of solving complex problems.
Quantum computers promise to solve problems that conventional computers cannot, because qubits can exist in multiple states at the same time. Using this quantum physics phenomenon, qubits can perform large amounts of calculations simultaneously, which can greatly speed up the speed of solving complex problems.
Traditional computer vs quantum computer
One of the differences between quantum computers and traditional computers is computing power. The former can solve a large number of operations that are difficult for traditional computers to handle.
For example, given the same complex data task, a quantum computer can do it in a matter of minutes, while today’s best-performing conventional computers would take thousands of years.
The key to this is the “qubits” that are the core of a quantum computer. From the perspective of quantum physics, qubits can exist in multiple states at the same time, and can perform a large number of operations several times higher than traditional computers, greatly speeding up the speed of solving complex problems.
For most quantum computers, qubits must be kept operating at extremely cold temperatures close to stopping atoms from moving. As a result, qubits are often placed in a special refrigerator, also known as a “quantum refrigerator,” while other devices are placed around the quantum refrigerator.
But controlling a quantum processor requires hundreds of wires to go in and out of the refrigerator, a wire design that would greatly constrain the ability of a quantum system to scale to the hundreds or thousands of qubits needed to demonstrate quantum utility, while also preventing Making qubits send and receive information is very difficult. This has also become a problem that scientists must solve in the process of advancing the development of quantum computing.
However, as companies managed to increase the number of qubits in a chip, and thus the computing power of the chip, they started to run into a problem.
The ultimate goal is to minimize the number of wires going into the chiller. Intel recognizes that quantum control is an integral part of the puzzle it needs to solve to develop large-scale commercial quantum systems.
Intel unveils cryogenic chips
At the IEEE International Electron Devices Conference in San Francisco this week, Intel Corp. unveiled a cryogenic chip designed to speed up a quantum computer they have developed in collaboration with the QuTech research group at Delft University.
The chip, called Horse Ridge, one of the coldest areas in Oregon, uses specially designed transistors to provide microwave control signals for Intel’s quantum computing chips.
The chip is designed to operate at 4 Kelvin, slightly above the temperature of the qubit chip itself. The company made the chip using its 22-nanometer FinFET process, although the transistors that make up the control circuitry required extensive redesign.
The chip can control multiple qubits in a quantum computer, and Intel sees the development of the chip as a major milestone on the road toward a truly viable quantum computer.
Intel claims that Horse Ridge lays the foundation for future controllers that will be able to control thousands or even millions of qubits, enabling the realization of quantum computers. Miniaturization is key, they claim, and it’s worth noting that miniaturization is one of Intel’s strong suits.
Dealing with the Difficulties of Quantum Computers
While most quantum chips and computers need to be placed at absolute zero to function properly, the Horse Ridge chip operates at about 4 degrees Kelvin, which is slightly warmer than absolute zero.
Because each of these particles is individually controlled, the wiring’s ability to scale quantum computing systems to hundreds or thousands of qubits achieves remarkable performance levels.
Horse Ridge SoCs use sophisticated signal processing techniques to convert instructions into microwave pulses that can manipulate the state of qubits.
The solution is to put as much control and readout electronics as possible into the refrigerator, possibly even integrating them on a qubit chip.
Horse Ridge integrates the control electronics onto a chip used to operate inside the refrigerator using the qubit chip. It is programmed with instructions corresponding to basic qubit operations, which are converted into microwave pulses that can control the state of the qubits.
Milestones of Horse Ridge
For a long time, in the race to realize the functions of quantum computers and unleash their potential, researchers have paid more attention to the manufacture of qubits, building test chips to demonstrate the powerful capabilities of a few qubits in superposition states.
But early quantum hardware development at Intel, including the design, testing and characterization of silicon spin qubit and superconducting qubit systems, identified the main bottlenecks preventing quantum computing from scaling commercially: interconnect and control.
Intel sees Horse Ridge as opening an “elegant solution” that allows multiple qubits to be controlled, and sets a clear path for building systems that can control more qubits in the future, an important milestone toward quantum utility .
Through Horse Ridge, Intel is able to scale quantum systems to the hundreds or thousands of qubits needed to demonstrate quantum utility, and thus the millions of qubits needed to achieve commercially viable quantum solutions.
Manufacturing the control chip, which is done in-house at Intel, will greatly improve the company’s ability to design, test and optimize commercially viable quantum computers.
These devices are often custom-designed to control individual qubits, requiring hundreds of connecting wires to go in and out of the refrigerator. But this extensive control cable for each qubit hinders the ability to scale quantum systems to the hundreds or thousands of qubits needed to demonstrate quantum utility, let alone commercially viable quantum solutions. Millions of qubits.
With Horse Ridge, Intel can radically simplify the control electronics needed to operate quantum systems. Replacing these bulky instruments with highly integrated SoCs will simplify system design and allow the use of sophisticated signal processing techniques to speed up setup times, improve qubit performance, and enable systems to efficiently scale to larger qubit counts.
Microsoft and Amazon join the fray
Quantum computing is a hot research field. Although we have not yet seen what a quantum computer is, IBM, Google, and Amazon are already vying for the quantum market.
Quantum computing will enable most consumption to be realized through the cloud. If it can be successful, quantum computers will have a very amazing increase in computing power. Judging from the prototype pictures disclosed by various companies, they are all huge and controlled by hundreds of wires. A behemoth with dense lines.
Of course, other quantum computing companies with massive qubit numbers are working on the same problem. Earlier this year, Google described some ideas for a cryogenic control circuit for its machines. In short, Intel’s breakthroughs are very helpful for them to launch higher-integrated quantum chips.
・Microsoft announced a cloud computing service called Azure Quantum at its Ignite conference in November; it integrates Microsoft’s previously released quantum programming tools with cloud services, allowing coders to work on simulated quantum hardware or on real quantum computers Run quantum code.
・Amazon unveiled a preview of Amazon Braket at AWS re:Invent in December; it also said that the creation of the “AWS Quantum Computing Center,” a physics lab near Caltech, is underway Bringing together the world’s leading quantum computing researchers and engineers to accelerate the development of quantum computing hardware and software.
Quantum computing currently faces many challenges, one of which is that superconducting qubits only really work at temperatures close to absolute zero.
Both Google and IBM required bulky control and cooling systems to develop quantum computing, some tubes larger than a human being, and hundreds of wires to connect to external microwave transmitters.
Despite the great emphasis placed on the qubits themselves, the ability to control multiple qubits simultaneously has been an industry challenge. Intel recognizes that quantum control is one of the most pressing challenges for us to develop large-scale commercial quantum computing systems. That’s why we’re investing in quantum error correction and control.
Competitors to Intel, Google and IBM, are primarily focused on superconducting qubits, the quantum computing systems driven by them that need to operate in the millikelvin range, just a tad above absolute zero.
But Intel believes that silicon spin qubits have the potential to work at higher temperatures, about 1 Kelvin, in hopes of enabling differentiated competition.
Given that Intel once tried to recreate the leadership of computer chips in the field of mobile chips, poured years of effort and huge sums of money, but it still ended in failure, now calling Horse Ridge a disruptive achievement and a “killer” that surpassed Google and IBM. It’s too early.