Now in preview, Amazon Braket is a new service AWS will be offering to make it possible to build, test, and run quantum algorithms. Braket includes a development environment, support for testing quantum algorithms on simulated quantum computers, and the ability to run them on existing quantum processors.
Braket development environment uses Python and provides a notebook-style interface. It is based on the Amazon Braket SDK, which is currently only available on AWS Cloud.
As mentioned, Braket enables running your quantum algorithms on quantum processors from different companies, including D-Wawe, IonQ, and Rigetti. Testament to Quantum computing being still in its infancy, those three quantum processors adopt radically different approaches to quantum computing.
D-Wave system uses quantum annealing instead of the gate-model quantum computer approach that is being pursued by Google, Microsoft, IBM and others. Quantum annealing allows D-Wave to claim a higher qubit number for its processors, up to 2048 indeed. In comparison, Google’s processor, known under the code name of Sycamore, has “only” 53 qubits. It is clear, though, that the two concepts of qubits cannot be compared directly. In fact, Google recently claimed to have proved quantum supremacy using Sycamore, while simulated annealing processors are usually used for a reduced class of problems defined in a discrete search space with many local minimums.
Rigetti Computing is attempting to build a quantum architecture that lend itself to rapid scaling. Thanks to this, Rigetti is aiming to create a 128 qubit processor.
IonQ uses an approach based on a trapped ion architecture which was presented in a IEEE paper two years ago. In short, IonQ architecture aims to make it possible to create quantum computers able to operate at room temperature leveraging the technological foundations of atomic clocks. IonQ further claims their approach is scalable and enables the construction of quantum computers that can target any type of quantum application.
In actuality, quantum processors still require very stringent operational conditions, for example super cooled environments free of electrical, thermal, and magnetic noise. This makes the Cloud the ideal means to provide access to them to a larger public, says AWS Chief Evangelist Jeff Barr.
I think it is safe to say that most organizations will never own a quantum computer, and will find the cloud-based on-demand model a better fit.
Most comments on Hacker News focused rightly on this aspect of the offering of a Quantum Cloud, comparing this to the time sharing era, when owning a computer was not achievable for many organizations, let alone individual users.
It is worth noting that Amazon is not operating all of those quantum processors in its own facilities. Instead, it will just provide a convenient and integrated access to the cloud services operated by their respective owners. This also makes it possible to extend the set of quantum processors supported by including additional existing quantum Cloud services.
As a final note, Amazon Braket is not the only Cloud Computing as a Service platform currently available. Most notably, both IBM and Microsoft have also been pushing their Quantum Clouds forward. IBM has been offering its IBM Q Cloud for a while now, while Microsoft recently introduced Azure Quantum, which follows an approach similar to Braket’s making different quantum devices available through a unified stack.
.(tagsToTranslate)amazon braket quantum cloud(t)Development(t)AWS(t)Amazon(t)Quantum Computing(t)IaaS(t)Cloud Computing(t)Cloud(t)Emerging Technologies
This is a syndicated post. Read the original post at Source link .