NVIDIA’s first virtualized GPU announcement, 2012
Jensen Huang announces NVIDIA’s plan to bring Kepler GPUs into cloud data centers. He describes three enabling features: GPU virtualization so multiple users can share hardware, rendering without a directly attached display, and energy efficiency sufficient for large-scale deployment. Huang then explains how virtualized graphics command buffers identify each virtual machine and return its rendered frame buffer remotely.
Passages
1:15 “Kepler is the world’s first GPU designed for the cloud. Kepler is the world’s first GPU we can deploy into cloud data centers all over the world.” Share
1:45 “For the first time, we have virtualized the gpu. Second, no longer does this GPU need to directly connect to a display.” Share
Search what Huang said from Claude, ChatGPT or your own code. See plans