NVIDIA’s latest supercomputer announcement is aimed at reducing the time it takes to deploy huge AI number crunching systems to a matter of weeks. The NVIDIA DGX SuperPOD™ Solution for Enterprise is sold as up to seven individual 20-unit clusters of NVIDIA DGX A100™ systems interconnected with NVIDIA Mellanox® HDR InfiniBand networking and using all-flash storage. Performance wise, a SuperPOD solution can range from 100 to 700 petaflops.
“Traditional supercomputers can take years to plan and deploy, but the turnkey NVIDIA DGX SuperPOD Solution for Enterprise helps customers begin their AI transformation today. State-of-the-art conversational AI, recommender systems and computer vision workloads rapidly exceed the capabilities of traditional infrastructure, and our new solution gives customers a fast track to the world’s most advanced, scalable AI infrastructure and NVIDIA expertise.”
Charlie Boyle, VP and GM, DGX systems, NVIDIA
NVIDIA Selene (see video above) is a 280-node DGX SuperPOD located in the U.S. and the recently announce Cambridge-1 supercomputer, which will be used for AI and healthcare research, will be built as a 80-node DGX SuperPOD with 400 petaflops of AI performance. Solutions are already deployed in Korea, Sweden and India.