Nvidia is adapting to both AI and improvements needed in data center GPUs for non-AI work. View a comparison of their GPUs here.
Category: GPUs
Virtualization has revolutionized the way we use computer resources. One particular element is virtual GPU (vGPU) that has ability to deliver high-performance graphics and accelerate complex tasks. vGPU has become indispensable in industries like desktop virtualization (VDI) and remote workstations, ML/AI workloads, and scientific research. Within OpenStack clouds, the project Nova acts as a bridge between physical GPUs and the VMs that need GPU resources. Nova efficiently manages and allocates virtual GPUs. In this blog, we will explore Nova and vGPUs, their practical applications, and the process of setting up vGPUs with OpenStack Nova.
In this video and accompanying article OpenMetal’s Principal Engineer, Jacob Hipps, discusses various use-cases and considerations for utilizing vGPUs effectively in OpenStack Nova. This session was recorded at OpenInfra Summit 2023 in Vancouver.