Stanford University, in partnership with NVIDIA and Mark III Systems, has deployed Marlowe, a DGX SuperPOD equipped with 248 NVIDIA H100 GPUs, giving more than 500 researchers across all seven of its schools access to large-scale computing. It is the university's first GPU-based supercomputer, intended to support interdisciplinary work in fields such as materials science, neuroscience, medical imaging and AI. A NVIDIA case study and the Stanford project page detail the system.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.