Meta (erstwhile Facebook) is the latest company to build an AI supercomputer. Meta?s AI Research SuperCluster has been designed to train machine learning systems and the company says that it will be the world?s fastest supercomputer once it is complete later in 2022. The new supercomputer was announced by Meta CEO Mark Zuckerberg, who said that it will be completed later this year. ?Meta has developed what we believe is the world?s fastest AI supercomputer. We?re calling it RSC or AI Research SuperCluster and it?ll be complete later this year,? Zuckerberg said in a statement.
The supercomputer reflects the heavy investments big tech companies are making in AI research. Other tech companies like Microsoft and Nvidia have also announced their own ?AI supercomputers.? These are slightly different from normal supercomputers, reports have suggested. The AI Research SuperCluster from Meta will be used to train a range of system?s across the company?s businesses. This includes algorithms on content moderation for detecting hate speech and explicit content on Facebook and Instagram, to even AR programmes that will be available in Meta?s AR gadgets in the future.
ALSO READ: Meta To Educate People About Data Privacy At New Privacy Centre
As one would expect, the RSC will be used to design things or run programs for the Metaverse. In a blog post, Meta engineers Kevin Lee and Shubho Sengupta say that the RSC will help build the company?s AI models that can learn from ?trillions of examples and work across hundreds of different languages. It will also seamlessly analyse text, images, and video together, develop new reality tools, and much more.?
Meta started working on the AI Research SuperCluster about a year and a half ago, with the company?s engineers designing the machine?s components for cooling, power, networkin, and cabling, entirely from scratch.
ALSO READ: Here?s How Instagram Is Reducing Potentially Harmful Content On the Platform
The Supercomputer consists of 760 Nvidia GGX A 100 systems containing 6,080 GPUs. Meta says that it is already providing up to 20 times improved performance. It will have a total of 16,000 GPUs by the end of the second phase of its creation. This is expected to happen before the end of 2022.
Now, AI supercomputers are different from normal computers in the sense that machine learning requires less accuracy than the kind of tasks handled by traditional supercomputers. Hence, ?AI Supercomputers? can carry out more calculations per second than their regular counterparts using the same hardware. This basically means that the Meta AI Reasearch SuperCluster is not a direct competitor to supercomputers like the Summit 500 or Fugaku, which is currently the fastest supercomputer in the world. While a traditional supercomputer?s performance is measured in 64-bit floating point operations per second (FLOPs), an AI Supercomputer is often measured in 32-bit or 16-bit FLOPs.
Read all the Latest Tech News here
Comments
0 comment