Microsoft Develops Cryo-Controller Chip – Gooseberry – for Quantum Computing
Computing

Microsoft Develops Cryo-Controller Chip – Gooseberry – for Quantum Computing

[ad_1]

Researchers from Microsoft and the University of Sydney, Australia, have reported developing a cryo-controller chip – named Gooseberry – and subsystem for quantum computing devices. Their work, published in Nature Electronics this month, tackles the tricky challenge of how best to control qubits which must typically be housed at near-zero (K) temperatures inside dilution refrigerators.

The current practice is to cram coaxial cables from a room temperature environment housing the control electronics into the dilution refrigerator to the quantum processor, literally connecting one cable to each qubit. Not only is this unwieldy and chancy – a broken cable connection turned Google’s 54-qubit device into a 53-qubit system during its work to demonstrate quantum supremacy – but also it limits the scalability of quantum computers generally.

Accommodating 50 or so qubits, which represents the top end of NISQ (noisy intermediate-scale quantum) computers today, is one thing; packing in hundred, thousands, or even more cables necessary to scale up to implement fully error-corrected quantum devices is quite another task. Microsoft isn’t the only company to tackle cryo-controllers. Intel has also developed a cryo-controller chip – Horse Ridge 2 – targeting use with its CMOS quantum dot spin qubit technology (see HPCwire coverage).

The Microsoft work also is CMOS-based, which the researchers note should allow scaling up. Among other things, the researchers implemented an effective heat dissipation scheme so the CMOS chip could be in close proximity to the qubits (quantum chip) without disrupting the fragile quantum states on which quantum computing depends. Microsoft tested the chip on a CMOS-based silicon dot quantum processor; it is investigating several qubit technologies and has received most notice for work on so-called topological qubits based on Marjorana particles, which if it works, would eliminate many error correction issues.

In their paper (A cryogenic CMOS chip for generating control signals for multiple qubits), the researchers write:

“Our CMOS chip is a 2.5 mm X 2.5 mm integrated circuit with around 100,000 transistors. A serial peripheral interface (SPI), which consists of four low-bandwidth wires connected at room temperature, is used to provide the digital instructions (input signals) to the chip. These input signals are handled by the digital logic of an on-chip finite state machine (FSM), which then configures 32 analog blocks, each of which can be used to control a single gate of a qubit. These analog circuit blocks – termed charge-lock fast-gate (CLFG) cells – use the low leakage of the transistors at cryogenic temperatures to store and shuffle charge between the floating capacitors to generate the dynamic voltage signals for manipulating qubits.

“Compared with a direct connection to room temperature or 4K via cable, moving the stored charge between on-chip capacitors consumes significantly less power and has a smaller footprint of 100 micrometers X 100 micrometers for a single CLFG cell. We benchmark this architecture on a GaAs few-electron quantum dot (QD) device…[The] results suggest that complex circuits based on CMOS technology can be designed to operate near 100mK and can potentially provide a scalable platform for controlling the large number of qubits needed to realize quantum applications.”

There’s also an account of the work in a Microsoft research blog by Chetan Nayak posted yesterday. Here’s an excerpt:

“Microsoft Quantum researchers are playing the long game, using a holistic approach to aim for quantum computers at the larger scale needed for applications with real impact. Aiming for this bigger goal takes time, forethought, and a commitment to looking toward the future. In that context, the challenge of controlling large numbers of qubits looms large, even though quantum computing devices with thousands of qubits are still years in the future.

“Enter the team of Microsoft and University of Sydney researchers, headed by Dr. David Reilly, who have developed a cryogenic quantum control platform that uses specialized CMOS circuits to take digital inputs and generate many parallel qubit control signals—allowing scaled-up support for thousands of qubits—a leap ahead from previous technology. The chip powering this platform, called Gooseberry, resolves several issues with I/O in quantum computers by operating at 100 milliKelvin (mK) while dissipating sufficiently low power so that it does not exceed the cooling power of a standard commercially-available research refrigerator at these temperatures. This sidesteps the otherwise insurmountable challenge of running thousands of wires into a fridge.”

Shown below are a pair of figures from the paper.

 

Qubit control is one of the thornier obstacles for modern quantum computers requiring low temperature environments. The Microsoft work as well as Intel’s is significant. Many observers think it will take systems with millions of qubits to implement wide-spread practical quantum computing. Many efforts are underway to find ways to scale up. IBM, for example, has announced a project – Goldeneye – to develop a 10-foot-tall and 6-foot-wide “super-fridge” able to house a million-qubit system.

One wonders whether Intel, traditionally a component supplier, might supply versions of its cryo-controller chip to the emerging quantum computer systems market. Intel is also working on CMOS-based, quantum dot technology for use as qubits. Microsoft’s quantum foray is interesting in that it is exploring various qubit technologies. Noteworthy, Microsoft Azure offer Azure Quantum, a portal that provides access to several quantum computers from different vendors plus various tools.

Link to Nature Electronic paper: https://www.microsoft.com/en-us/research/publication/a-cryogenic-cmos-chip-for-generating-control-signals-for-multiple-qubits/

[ad_2]

Source link

Spread the love

Leave a Reply

Your email address will not be published. Required fields are marked *