pve 8 + Nvidia P4 + vgpu ，Hign

[kkek]

I'm experiencing a significant increase in memory usage on the PVE host when using PVE with Nvidia P4 and vGPU. On average, each virtual machine with a 1Q vGPU results in an additional 4GB of memory consumption.

 

[dakralex]

When a guest uses GPU passthrough (or any PCI passthrough for that matter), it will allocate main memory for the IOMMU group that was assigned to it so that the virtual machine can directly communicate with the device as if it was physically connected. But since you use a P4-1Q vGPU profile it seems unusual that it will allocate 4GB of main memory for 1GB of graphics memory. How did you set up your PVE and your guests?