Meta unveils the AI ​​Research SuperCluster, an artificial intelligence supercomputer that will eventually be the most powerful in the world

Mark Zuckerberg, founder of Facebook and CEO of Meta, announced Monday, January 24, the construction of the supercomputer AI Research SuperCluster (RSC), which is among the fastest AI supercomputers today and should be the fastest by mid-2022, when finalized. Researchers are already using it to train large models in natural language processing (NLP) and computer vision. RSC will enable the creation of technologies for the Metaverse, the next great computing platform where AI-driven applications and products will play a significant role.

The first calculators appeared in the 1960s and were first used for scientific research. Since then, their power has continued to increase and we now measure their computing capacity in exaflops (more than a billion billion operations per second). The United States, China, Japan and Europe are in the race for exascale computers with the challenge of world technological domination. RSC will allow Meta to make a name for itself in the field of metaverse, an ambition clearly affirmed when Facebook changed its name to Meta:

Mark Zuckerberg clarifies about RSC:

The Research SuperCluster supercomputer

The supercomputer will allow Meta to use data from its various platforms (Facebook, Messenger, Instagram, etc.) to train AI engines, particularly in the field of language or image processing. It will replace Meta's infrastructure designed in 2017, which has 22,000 Nvidia V100 Tensor Core GPUs and performs 35,000 training jobs per day.

He also assured:

A super-powerful supercomputer

Supercomputers are built by combining multiple GPUs into compute nodes which are then connected by a high-performance network fabric that enables rapid communication between GPUs. RSC includes a total of 760 Nvidia DGX A100 systems as compute nodes, for a total of 6,080 GPUs. Each GGX communicates through an NVIDIA Quantum 1600 Gb/s bi-tier Clos fabric with no oversubscription. RSC's storage tier includes 175 petabytes of Pure Storage FlashArray, 46 petabytes of cache memory in Penguin Computing Altus systems, and 10 petabytes of Pure Storage FlashBlade. fifth supercomputer in the world in terms of computing power, according to the TOP500 list which is a benchmark in this area. Meta continues to evolve RSC with the aim of making it the most powerful supercomputer in the world, particularly in terms of the cache and storage system, which should reach 1 exabyte, or the equivalent of 36,000 years of high quality video. It will become "to our knowledge the most powerful artificial intelligence supercomputer in the world", assures Meta.