Ocelot, the New Quantum Computing Chip from AWS

Ocelot is the latest development from Amazon Web Services (AWS). It is a prototype chip that tests a quantum error correction architecture.

One of the biggest challenges facing the industry when building quantum computers is their high sensitivity to changes in the environment, including vibrations, heat or electromagnetic interference that cause computational failures.

To address this, AWS has developed Ocelot from the ground up with built-in error correction.

To suppress certain forms of error and limit the amount of resources needed for error correction, AWS uses the cat qubit, named after Schrödinger’s famous experiment, and combines it with additional quantum error correction components.

Ocelot is estimated to be able to reduce implementation costs for such correction by 90 %.

The AWS proposal has a pair of silicon microchips, on the surface of which are superconducting materials. Its circuits have 5 data qubits, 4 qubits for detecting errors in the cat qubits and 5 buffer circuits.

Ocelot was developed by the team at the AWS Centre for Quantum Computing and the US company is committed to investing further to optimise its approach. The researchers have published a paper in Nature.

The approach applied to Ocelot is expected to enable smaller, more reliable quantum computers to be built at lower cost.

This should lead to applications such as the production of new materials, drug discovery, accurate risk predictions and other problems beyond the reach of current machines.

‘With recent advances in quantum research, it’s no longer a question of ‘what if’, but ‘when practical’, fault-tolerant quantum computers will be available for real-world applications,’ says Oskar Painter, director of Quantum Hardware at AWS.

‘Ocelot is an important step on that journey,’ he notes. ‘In the future, quantum chips built according to the Ocelot architecture could cost as little as one-fifth of today’s approaches, due to the drastically reduced amount of resources needed for error correction.’

‘Specifically,’ he says, ’we believe this will accelerate our timeline to a practical quantum computer by up to five years’.

Painter explains that ‘we looked at how others were approaching quantum error correction and decided to take a different path’.

‘We didn’t take an existing architecture and then try to incorporate error correction into it,’ he says. ‘We chose our qubit and our architecture with quantum error correction as the main requirement.

‘We believe that if we are going to make practical quantum computers, quantum error correction must come first,’ he insists.

He estimates that scaling up Ocelot to a ‘full quantum computer capable of transformative societal impact would require only a tenth of the resources associated with standard quantum error correction approaches’.