Is it possible to transfer Nvidia GPU graphics profiles between vms?
Article ID: 379573
Updated On:
Products
Issue/Introduction
This article explains about the possibility of sharing Nvidia GPU graphics profiles between virtual machines.
Environment
VMware vSAN 7.x
VMware vSAN 8.x
VMware vCenter Server 7.x
VMware vCenter Server 8.x
Resolution
The number of physical GPUs that a board has depends on the board.
Each physical GPU can support several different types of virtual GPU (vGPU).
Each NVIDIA vGPU has a fixed amount of GPU framebuffer, and one or more virtual display outputs or "heads".
The vGPU’s framebuffer is allocated out of the physical GPU’s framebuffer at the time the vGPU is created, and the vGPU retains exclusive use of that framebuffer until it is destroyed.
Conclusion: "Dedicated GPU Memory" within Guest OS indicates a fixed amount of GPU frame buffer and it is reserved for each VM instance.
Additional Information
Refer to the Nvidia document for more details: Virtual GPU Software User Guide and the following reference image is also from the same document which explains about dedicated framebuffer for each VM.
Feedback
