Microsoft reveals immense cloud supercomputer, just for OpenAI

AMD chip

At its now-virtual Build conference, Microsoft detailed the cloud supercomputer it is running for AI research group OpenAI.

Last year, the technology giant announced it would invest $1bn in the group, which essentially shed its non-profit status as it pushes into commercial operations.

Microsoft claims that the supercomputer would be the fifth fastest system in the Top500 supercomputer rankings.

It’s a big one
Microsoft said it built the system in collaboration with OpenAI, and that the supercomputer had a whopping 285,000 CPU cores, 10,000 GPUs and 400 gigabits per second network connectivity for each of the GPU servers.

“As we’ve learned more and more about what we need and the different limits of all the components that make up a supercomputer, we were really able to say, ‘If we could design our dream system, what would it look like?’” OpenAI CEO Sam Altman, said. “And then Microsoft was able to build it.”

Microsoft CTO Kevin Scott added: “The exciting thing about these models is the breadth of things they’re going to enable.”

“This is about being able to do a hundred exciting things in natural language processing at once and a hundred exciting things in computer vision, and when you start to see combinations of these perceptual domains, you’re going to have new applications that are hard to even imagine right now,” he said.

The machine will be used for OpenAI’s long term research into artificial general intelligence, as well as for nearer term goals of developing Azure AI supercomputing technologies, and other AI tools – with Microsoft the ‘preferred commercialization partner.’

Source: datacenterdynamics.com