DOES Artificial Intelligence Need Special Computers ?
In this presentation the extreme scaling capabilities within a single DGX A100 node with 5 Petaflop/s (Tensor Core operations) will be demonstrated. Following that the mature and advanced DL and ML ecosystem supporting DGX will be highlighted. Some architectural blueprints will be concluded for massive multi-node scaling to further expand overall performance and provide an inside view on the SaturnV AI-Supercluster.
Add Event to Calendar