Amazon AWS: Amazon and Nvidia to build the world’s largest cloud AI supercomputer

Amazon Web Services has revealed that it is in partnership with GPU-maker Nvidia to launch the Project Ceiba supercomputer. This supercomputer is expected to be the world’s largest cloud AI supercomputer, featuring  16,384 GH200 NVL32 Grace Hopper Superchips. The system will also feature the AWS Elastic Fabric Adapter (EFA), the AWS Nitro system, and Amazon EC2 UltraCluster scalability technologies.

Ian Buck, Nvidia's VP of hyperscale and HPC, said, “This new supercomputer will be set up inside of AWS infrastructure hosted by AWS and used by Nvidia’s own research and engineering teams.” It will be used "to develop new AI for graphics, large language model research, image, video, 3D, generative AI, digital biology, robotics research, self-driving cars, and more.”

Amazon said that the supercomputer is expected to provide 65 exaflops of AI performance and have up to 9.5 petabytes of total memory.