What if you had a computer that could handle more than two hundred quadrillion calculations per second, or roughly five hundred thousand times the number of stars in the Milky Way? And, could store the entire world’s social media data in one place? What if that super computer was already deployed and beginning to run test cases?
Over the past years, IBM, NVIDIA, Mellanox, and Red Hat all worked together to create Summit and Sierra for Oak Ridge (ORNL) and Lawrence Livermore (LLNL) National Labs. These state-of-the-art supercomputers are built with everyday IBM POWER9™ processors and IBM Storage that will run traditional HPC workloads and AI workloads. They were not developed as highly customized, one-off systems seen in previous HPC solutions. Rule number one: whatever was built for ORNL and LLNL needed to work for the enterprise, as well.
Summit is the most powerful and smartest supercomputer for science. The architecture delivers massive HPC scalability on a system purpose-built for AI. “By building these supercomputers, we are building the world’s leading AI machines,” says Hillery Hunter, IBM Fellow; Director, Accelerated Cognitive Infrastructure.
Researchers will use the computers to simulate quantum mechanics materials, explore evolutionary relationships based upon genomes, and fabricate nanomaterial using AI.