AWS and QuEra Computing Partnership Brings Libra Fault Tolerant Quantum Computing to Amazon Braket Cloud for Scientific Research
The ongoing strategic partnership between AWS and QuEra Computing is poised to bring fault tolerant quantum computing directly to the cloud. As the collaboration looks to make the new Libra quantum processor accessible via Amazon Braket, a shift towards practical compute from theoretical research for complex problems in physics and chemistry is on the horizon by the end of the decade.
The 2 companies established this collaboration over 5 and a half years ago, leading to the launch of Aquila, a 256 qubit analog Rydberg device built for academic and industry research. Since the launch of the first 256 qubit processor, a variety of research fields including quantum reservoir computing, high energy physics simulations, and financial optimization have explored its capabilities. Libra aims to be the next step, pushing beyond primitive quantum computer architectures towards highly reliable compute.
"This is a very special moment, for the 1st time, a dream of realizing useful, fault tolerant quantum computers is in our direct line of sight. Designed to enable quantum computation at an unprecedented scale, these systems should realize truly unique applications. We are proud to significantly expand our collaboration with AWS to bring these unique capabilities to the broader community of scientific users."
These words from Professor Mikhail Lukin, the Chief Science Officer at QuEra Computing illustrate the movement towards commercially viable quantum computer hardware. Libra, or the Megaquop scale device will be capable of performing 1,000,000 quantum operations on 100s of logical qubits and will be able to handle material science, molecular chemistry, and high energy physics problems too large for classical supercomputers.
There are 2 fundamental physical advantages to using neutral atom quantum computers: First, they are physically scalable, allowing the construction of dense qubit arrays on the scale of 10,000s to 100,000s active qubits on a single chip, meaning there is no immediate need for inter chip connections. Secondly, using optical tweezers, individual atom qubits can be repositioned within the array without losing their coherence states, which greatly increases the efficiency of error correction through all to all connectivity.
The AWS platform, Braket, is designed to provide users with an easy to use hybrid application development environment that will provide access to AWS’s classical computing capabilities, as well as popular open source tools including Qiskit, PennyLane, Bloqade, and CUDA Q. Due to the nature of early fault tolerant applications, which rely on tight co design between the hardware and the algorithms that utilize it, AWS and QuEra are working to provide customers and their software partners with support in developing applications from the ground up.
Along with the development of the neutral atom systems via the partnership with QuEra, AWS is simultaneously developing their own proprietary hardware. At the AWS Center for Quantum Computing, engineers are building Ocelot, a superconducting chip designed around the cat qubit architecture. This dual strategy follows the principle learned from 20 years of developing cloud infrastructure—a single chip design cannot be ideal for every task. The fundamentally fast clock rates of superconducting chips make them suitable for deep circuit calculations, while the extremely dense qubit array on neutral atom systems such as Libra offer scalability through physical arrangement and re arrangement of qubits.
This latest system’s launch onto the Amazon Braket platform demonstrates a clear commitment towards realizing practical quantum advantage. Rather than being an isolated computing resource, these quantum computers will be utilized as part of the AWS cloud, in much the same way as CPUs and GPUs serve as specialized computing resources. Customers interested in preparing their software pipelines for these future fault tolerant devices can engage directly with the hardware teams.
