Nvidia Works on Meta's AI Supercomputer to Power Metaverse
Boris Ocić, Dražen Tomić 29 Jan 2022 Print Comment
Meta chose Nvidia's technologies for a new supercomputer designed to scale up and develop the next generation of AI technology. Meta will use it to accelerate its vision of the metaverse.
The company said that the AI Research Supercluster (RSC) will be the fastest supercomputer in the world once fully built by mid-2022, and would advance AI beyond current tasks such as translating text between languages and helping to identify harmful content. Through RSC, Meta Platform’s AI researchers are looking to develop models which can learn from trillions of examples, such as working across hundreds of languages, analyzing text, images, and videos together, and developing AR tools.
“As RSC moves into its next phase, we plan for it to grow bigger and more powerful, as we begin laying the groundwork for the metaverse,” Meta stated. The new AI supercomputer currently uses 760 NVIDIA DGX A100 systems as its compute nodes. They pack a total of 6,080 A100 GPUs linked on a Quantum 200Gb/s InfiniBand network to deliver 1,895 petaflops of TF32 performance.
More from category
IBM to Build its First European Quantum Data Center
IBM announced plans to open its first Europe-based quantum data center to facilitate access to cutting-edge quantum computing for companies, research institutions, and government agencies.
SoftBank to Use Nvidia Superchips in Data Centers
SoftBank unveiled plans to power future data centers in Japan with Nvidia’s latest Arm-based AI chip.
NVIDIA CEO Unveils Gen AI Platforms for Every Industry
With his first live keynote since the pandemic, NVIDIA founder and CEO Jensen Huang kicked off the COMPUTEX conference in Taipei.