Your Browser is Out of Date

ShareDemos uses technology that works best in other browsers.
For a full experience use one of the browsers below
Citrix Virtual Apps and Desktops VDI Reference Architecture for Dell EMC PowerFlex family

close

view all chapters

Best practices
Management infrastructure Windows 10 optimization Scaling guidelines NVDIA vGPU considerations

NVDIA vGPU considerations

NVDIA vGPU considerations

  • NVIDIA vGPU supports various scheduling policies that allow VMware vSphere Shared Direct (passthrough graphics) feature to share the vGPU resources optimally with multiple workload VMs.

    • Best effort scheduler – default policy in vSphere, allows a vGPU virtual machine to use the GPU processor cycles are not used by other vGPUs virtual machines.
    • Equal share scheduler – shares the GPU processing equally to all the vGPU virtual machines are running.
    • Fixed share scheduler – each vGPU has the fixed share of the GPU allocated to the virtual machine.

    Equal share scheduler or Fixed share scheduler is recommended for VDI solutions, which reduces the overall CPU overhead to handle the vGPU scheduling for the virtual machines running in the ESXi host and improves the end-user graphics performance equivalently.