p7jsuh
Contributor
Contributor

ESX VM Video Memory settting

Jump to solution

Hi all,

Some basic high level questions I just can't seem to find the answer to. 

When editing the settings of a VM from the ESX console, there is an option to increase Video Memory.

1) What physical resource does this come from?  The bare metal host's RAM?  The host's dedicated video memory?  I'm aware there is the ability to passthrough a GPU, but this pertains to a bare metal host running only onboard video card.

2) How can I monitor a VM's video memory usage?  i.e. how do I prove out that the VM's VRAM is at capacity and needs more, like a dedicated GPU card?  GPU isn't show in task manager as it is on  physical machine.

2b.)  How can I see what the overall VRAM usage is, if it's at capacity or not?

Thank you for enlightening me!!!

0 Kudos
1 Solution

Accepted Solutions
sjesse
Leadership
Leadership

This is all asuming you don't have a server class gpu

1) What physical resource does this come from?  The bare metal host's RAM?  The host's dedicated video memory?  I'm aware there is the ability to passthrough a GPU, but this pertains to a bare metal host running only onboard video card.

The normal hosts memory

2) How can I monitor a VM's video memory usage?  i.e. how do I prove out that the VM's VRAM is at capacity and needs more, like a dedicated GPU card?  GPU isn't show in task manager as it is on  physical machine.

You  can't its part of the VM processes

2b.)  How can I see what the overall VRAM usage is, if it's at capacity or not?

In-guest tools like Liquidware labs may be able to do this, but I've only seen them do this with server gpus

Overall my question is if you need to check this for anything other than an intellectual exercise, you probably should have a server GPU. In general most servers don't need gpus to do there normal functions, the memory used to display a console or the windows interface doesn't use a lot and I've never seen an issue. Desktops are different since they build in gpu acceleration since they assume there is some sort of gpu included, even an onboard one like intel

View solution in original post

0 Kudos
4 Replies
NathanosBlightc
Commander
Commander

What's your video card (Product/Vendor/Model)?

Is it supported inside the vSphere environment?

Did you install the corresponding VGA driver in your virtual machine? What's the status of VGA if you check it inside the device manager?!

Please mark my comment as the Correct Answer if this solution resolved your problem
0 Kudos
p7jsuh
Contributor
Contributor

Thank you for your response, but my question is far more high level than your responsive questions are relevant to.  There are options to allocate Video RAM resources to a VM.  I asking where those resources come from, whether that's from the bare metal host's RAM or dedicated video RAM.  The answer to that is irrespective to drivers, guest OS listed devices, etc.

The corollary to my question is how can I monitor either VM Video RAM usage or the ESX host's video RAM usage in total.

0 Kudos
sjesse
Leadership
Leadership

This is all asuming you don't have a server class gpu

1) What physical resource does this come from?  The bare metal host's RAM?  The host's dedicated video memory?  I'm aware there is the ability to passthrough a GPU, but this pertains to a bare metal host running only onboard video card.

The normal hosts memory

2) How can I monitor a VM's video memory usage?  i.e. how do I prove out that the VM's VRAM is at capacity and needs more, like a dedicated GPU card?  GPU isn't show in task manager as it is on  physical machine.

You  can't its part of the VM processes

2b.)  How can I see what the overall VRAM usage is, if it's at capacity or not?

In-guest tools like Liquidware labs may be able to do this, but I've only seen them do this with server gpus

Overall my question is if you need to check this for anything other than an intellectual exercise, you probably should have a server GPU. In general most servers don't need gpus to do there normal functions, the memory used to display a console or the windows interface doesn't use a lot and I've never seen an issue. Desktops are different since they build in gpu acceleration since they assume there is some sort of gpu included, even an onboard one like intel

View solution in original post

0 Kudos
p7jsuh
Contributor
Contributor

Thanks for the informative response.

So just to give a sense of use case, this environment is intended to host a series of virtual desktops, i.e. VMs running Win10 on guest OS.  The physical desktops that have been P2Ved do have some requirements that necessitate an add on graphics card, e.g. multiple large monitors, high resolution, running multiple GUIs that refresh continuously.  With the current climate, we are moving away from physical machines (and physical office) so the goal is to consolidate all physical desktops onto a server and the guest hosts are to be accessed via RDP.

One heavily utilized guest host runs fine when I RDP to it, but the same machine is "slow" when its primary user connects to it.  My remote setup is three 24" monitors, while this power user has four 32" monitors.  This leads me to believe that the graphics memory is not sufficient.  I increased the video memory of the guest OS to 256MB and there is some improvement, but not enough. It is likely we will go the route of purchasing and setting up a GPU, but I don't have anything to conclusively point out that this is the bottleneck, which I would of course like to have.

0 Kudos