The Met Office has unveiled the 14th generation of supercomputer it’s using, which will be able to produce 14-day forecasts with a similar accuracy that’s currently available for seven-day forecasts.

The new supercomputer, which in fact runs in Microsoft’s Azure Cloud, aims to support efforts such as better weather forecasts over longer periods of time and advanced climate research.

While the weather agency—who use 200 to 300 terabytes of operational data a day—has utilised supercomputers for the last 60 years, this transition to a cloud-based system is a marked step change in its technological direction. Before this, the Met Office had almost always run its own supercomputers onsite.

Chris Ewen, who serves as Chief Information Officer at the Met Office, depicted what Microsoft is providing as “supercomputing for science as a service.” “Essentially there’s more science than we have the compute power to deliver, so it’s not a question of how much compute do you need, it’s how much you can get,” he said.

Penny Endersby, the CEO of the weather agency, also noted: “What’s different about this one, and particularly the whole journey with Microsoft, is that we have decided this is now a job for specialists, and the core business of the Met Office is helping you make better decisions, to stay safe and to thrive through providing fantastic forecasts.”

With a larger capacity of computing power, the Met Office’s teams can run more complex weather models via the cloud, while its scalability also means they can expand capacities for certain research projects on a case-by-case basis—meaning they don’t have to spin up any new and required infrastructure.

Recommended reading

• UK Satellite to Measure Earth’s Forests in 3D to Fight Climate Change
• Glasgow Uni to Teach Environmental Impact of Digital Tech
• Scottish Satellite Service to Tackle Environmental Disasters

After the transition to a new cloud-based supercomputer, how could the technology of the moment—artificial intelligence and machine learning—come into play for the weather agency?

While Ewen stated that “We don’t know yet how our CPU-based supercomputer services will be augmented with machine learning,” though “A lot of research is being done at the Met Office and elsewhere to find out.” “We’re already in the game, preparing our people,” he noted.