Microsoft Releases Preview of Azure Quantum Service
Microsoft announced this week that its Azure Quantum service is at the public preview stage.
The service is notable for breaking with typical binary compute operations in order to solve complex problems. Organizations might use Azure Quantum to solve problems in "chemistry, medicine, finance and logistics," Microsoft suggested.
"Developers, researchers, systems integrators, and customers can use it to learn and build solutions based on the latest innovations -- using familiar tools in the most trusted public cloud," Microsoft's announcement contended regarding Azure Quantum, although it mostly seems to be a still-developing tool for highly specialized research projects at this point.
The Azure Quantum service depends on a qubit instead of a binary bit as the basis for its operations. A qubit can exist in multiple states, unlike the strict on-and-off of binary computing. It's that multistate circumstance that accelerates operations on the quantum computing side.
Quantum computing requires the use of specialized hardware, kept at freezing temperatures. However, Microsoft provides a means for developers to test Azure Quantum on standard computers, which is done through emulation.
Azure Quantum Resources
Microsoft offers its open source Quantum Development Kit, which uses the Q# quantum programming language. It's also possible to create jobs on Azure Quantum using the Python SDK.
The Azure Quantum service is backed by actual specialized hardware, called "trapped-ion quantum systems." These systems, provided by partners Honeywell Quantum Solutions and IonQ, are designed to perform quantum computing operations.
Users of Azure Quantum can choose which hardware option to use. Here's how Microsoft's announcement characterized the differences between the two trapped-ion quantum systems available:
Honeywell's system leverages mid-compute measurement and qubit reuse, allowing developers to write quantum algorithms in uniquely impactful ways. IonQ's system offers a dynamically reconfigurable system for up to 11 fully-connected qubits that lets you run a two-qubit gate between any pair.
Microsoft and its partner 1Qbit offer optimization algorithms based on quantum principles, called "quantum-inspired optimization algorithms," that developers can test on "a range of silicon including CPU, GPU and FPGA." It's also possible to use so-called "classic optimization algorithms," where just some of them are based on "standard physics," a Microsoft "Intro" document explained.
Organizations using the Azure Quantum service are billed in a pay-as-you-go fashion, but users also need to have an Azure Storage account. In addition, Microsoft's document described having to pay for a "provider" billing cost, which apparently refers to the hardware option selected when using the service.
Microsoft has been working on quantum computing efforts for about 16 years, but it first showcased its efforts in a big way during the 2017 Microsoft Ignite keynote talk by CEO Satya Nadella.
That talk included a panel of experts who explained at the time that quantum computing is based on the behavior of "fractionalized electrons" instead of the on-and-off approach associated with traditional silicon-based processors. In that same year, Microsoft also introduced quantum computing tools for developers that integrate with the Visual Studio development environment.
Quantum computing is fairly complex, but it promises breakthroughs in computing powers. The difference between it and binary computing is that while a 100-bit binary system would deliver 100 bits of information, 100 qubits would deliver "1,267,650,600,228,229,401,496,703,205,375 individual pieces of information." Brien Posey, a Microsoft Most Valuable Professional, explained that concept in this Redmond article on the topic.
Kurt Mackie is senior news producer for 1105 Media's Converge360 group.