A fault-tolerant quantum supercomputer is a step closer, according to Microsoft. A new roadmap from the Microsoft Azure Quantum team outlines the six steps needed to achieve the goal of creating a machine that can solve some of the world’s toughest problems. Microsoft also unveiled a “quantum co-pilot” and “quantum elements,” a product that integrates HPC, artificial intelligence, and quantum to accelerate scientific discovery.
The roadmap builds on the first creation of topological qubits last year. These are considered a requirement for true fault-tolerant quantum computers, as they are less susceptible to noise and error than other forms of qubits. Microsoft claims to use particles known as Majorana topological states to form the basis of its qubits. “It’s like inventing steel, sparking the industrial revolution,” says Krysta Svore, vice president of advanced quantum development at Microsoft.
Microsoft claims that logical qubits, made up of many physical qubits, are necessary for a true quantum supercomputer. The more stable the qubits you start with, the easier it is to progress to supercomputing levels where you need fewer physical qubits per logical qubit. The company says it has tried all forms of qubits, including spin, send, gatimon and others, but none have been measured effectively.
Its engineers saw the need for a stable and scalable quantum supercomputer with “topological qubits”. Achieving this required a breakthrough in physics that had eluded scientists for nearly a century. Microsoft achieved this last year by creating matter and manipulating it into a topological state. In this case, qubits can be manipulated more easily, are more stable, and have a smaller footprint allowing for greater scale.
Microsoft’s next step on the road to a quantum supercomputer is to create “hardware-protected qubits,” also known as topological qubits, that have built-in error protection and can scale to support reliable qubits. Each of these qubits must be less than 10 microns to fit a million on a credit card-sized chip and be fully manageable.
Other steps include improving the quality of these hardware-protected qubits to enable entanglement and reduce error rates. Then the number of qubits must be increased and assembled into a programmable QPU. The final steps are working on flexibility and moving towards what Microsoft calls rQOPS, a new metric for tracking reliable quantum operations per second.
There’s no timeline on when this will be achieved, but Microsoft Azure says it expects the entire roadmap to be completed in “years, not decades.” so we can see it launched before the end of this decade.
Microsoft Quantum Elements: Three Stages of Quantum Evolution
Microsoft divides the quantum era into three phases, starting with the current NISQ, or noisy phase of error-prone, small-scale quantum computers. “The final breakthrough will come when institutions are able to precisely design new chemicals and materials using a quantum supercomputer,” the company says. To get there, the industry will follow a path similar to the development path of classical supercomputers, moving from vacuum tubes to transistors and then to large-scale integrated circuits.
The next step is flexibility, where quantum systems begin to operate on logical qubits that are reliable and stable enough to perform real computations. When this is determined, the final level will begin when quantum supercomputers evolve to a level and capabilities not possible with current or future classical supercomputers.
The goal is to “squeeze the next 250 years of progress in chemistry and materials science into the next 25 years,” Microsoft CEO Satya Nadella said, hinting at the amount of computing power possible with a supercomputer. quantum. Part of that will come through the announcement of Azure Quantum Elements, a combination of quantum computing, artificial intelligence, and high-performance computing to accelerate scientific research.
According to Azure, this allows developers and scientists to reduce research and development time and prepare for extended quantum computing. He claims that some customers see a particular chemical simulation speed up 500,000 times – putting a general strain on computing in a minute.
Artificial intelligence and quantum are closely linked
“Chemistry is in everything,” said Ansgar Schaefer, vice president of BASF, an early user of quantum elements. Schäfer leads the company’s quantum chemistry research and said, “To be able to improve products and processes, it’s really about understanding the chemistry behind them at a microscopic level. The more complex the challenge, the greater the computing power required. » [Azure Quantum Elements] It’s a tool that gives us the extra capacity needed to help develop entirely new research methods and increase efficiency and speed of development. »
Quantum Elements integrates Microsoft AI chemistry models and runs them in a hybrid quantum/HPC environment. It can also be directed using Copilot, Microsoft’s Core Model natural language artificial intelligence tool.
Copilot has already launched across Microsoft’s product family, including Windows, GitHub, and Microsoft 365. “Azure Quantum Copilot helps scientists use natural language to think through complex chemistry and materials problems,” the company explained. .
