Meta is about to build the world’s fastest AI supercomputer which is supposed to be integrated in the machine learning industry. It will be fully completed by the end of 2022.

In a post on Facebook, Mark Zuckerberg stated:

“Meta has developed what we believe is the world’s fastest AI supercomputer. We’re calling it RSC for AI Research SuperCluster.”

However, companies such as Microsoft and Nvidia have already announced their AI supercomputers.

In a statement which announced the news, the Meta engineers Kevin Lee and Shubho Sengupta said:

“RSC will help Meta’s AI researchers build new and better AI models that can learn from trillions of examples; work across hundreds of different languages; seamlessly analyze text, images, and video together; develop new augmented reality tools; and much more.”

They also hope that RSC will help them build entirely new AI systems that can, for example, power real-time voice translations to large groups of people, each speaking a different language, so they can seamlessly collaborate on a research project or play an AR game together.

The work on RCS began about a year and a half ago when Meta’s engineers designed the machine’s various systems such as cooling, power, networking, etc. The first half of it is already done and it is running and consists of 760 Nvidia GGX A100 systems containing 6,080 connected GPUs.

The work will be done by the end of this year. In a blog post, Meta’s engineers say that through 2022 they will be working on increasing the number of GPUs from 6,080 to 16,000 which is supposed to make the AI training performance better.

Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , , ,
Special Projects Editor