Penguin Computing Partners with Meta to Help Build Toward the Metaverse
Penguin Computing, Inc. a division of SGH (Nasdaq: SGH) and a leader in high-performance computing (HPC) focused on artificial intelligence and machine learning, today announced its role in providing AI-optimized architecture and managed services for the AI Research SuperCluster (RSC) — Meta’s cutting-edge AI supercomputer for AI research.
Today Meta announced via a blog post the introduction of the RSC, which Meta believes is “among the fastest AI supercomputers running today and will be the fastest AI supercomputer in the world when it’s fully built out in mid-2022.”
Penguin Computing, a long-time partner with Meta, provided the managed services and AI-optimized infrastructure for the RSC, which is comprised of 760 NVIDIA DGX A100 systems with more than 6,000 NVIDIA GPUs and 46 petabytes of cache storage—enabling what Meta believes is one of the fastest AI supercomputers currently in operation. Penguin’s powerful supercomputing platform also includes strong parameters to ensure data integrity and managed security. In its final build-out in mid-2022, RSC will utilize more than 16,000 NVIDIA GPUs and 1 exabyte of storage, which Meta believes will make it the largest AI supercomputer in the world.
Penguin and Meta started working on Meta’s first AI infrastructure in 2017, focused on advancing AI supercomputing and building toward the Metaverse.
“As one of the market leaders in HPC and AI, Penguin Computing is pushing the limits of AI-optimized architecture for ultra-scale environments in partnership with our customers,” said Thierry Pellegrino, senior vice president at SGH and president of Intelligent Platform Solutions. “We’ve been building customized HPC and AI solutions for more than 20 years so that our customers, like Meta, can rely on our expertise to help them advance their technologies.”
Another facet of Penguin Computing’s unique value is its strong relationship with NVIDIA and Pure Storage. Penguin worked with Meta’s operations team on hardware integration to deploy the cluster and set up major parts of the control plane. NVIDIA provided Meta with its AI computing technologies featuring cutting-edge NVIDIA DGX systems with NVIDIA A100 Tensor Core GPUs, as well as NVIDIA Quantum InfiniBand networking and software stack components like NVIDIA NCCL for the cluster.
Pure Storage provided Meta with a highly performant, robust, and scalable storage solution to power the AI Research SuperCluster. Penguin’s hardware and software expertise helped to unite contributions from Penguin, NVIDIA, and Pure Storage. Together, these three partners were key to supplying Meta with an optimized solution. The new RSC will enable Meta to begin laying the groundwork for the Metaverse.