Share GPU in dockers or VMs for Plex and Blue Iris


Recommended Posts

Hi guys

 

Trying to find an answer to this but I have failed so far. I want to figure out if I can share a GPU amongst dockers. 

 

The scenario:


I want to run Plex (which uses hardware transcoding) and Blue Iris IP Camera software which also uses hardware transcoding on either VMs or Dockers.

 

I would like to know if there is a possibility to share the GPU with 2 different dockers or 2 different VMs. 

I have been running these applications on a single Windows 10 machine so far and they share the GPU load from what I can tell. I want to migrate these to Unraid and make them isolated from each other in either dockers or VMs.

 

I plan to use a Quadro P2000 card for this.

 

If there is a possibility to share the GPU amongst dockers, then that is awesome and I could go about and set it up that way. Alternatively, if the GPU cant be shared, I could always have a single VM, slap windows on it and run both Plex and Blue Iris on the same VM so they share the GPU as before. 


Thoughts and advice would be highly appreciated.

Link to comment

you can use a gpu simultan either

 

unraid (dockers, also more then 1 like plex, emby, ... at the same time)

or each VM

 

means, 1 gpu for docker(s), a sep. GPU for each running VM

 

sample usecase 3 gpu´s

- intel igpu for dockers (plex, emby, tvheadend, ....)

- Nvidia GT1030 for VM1 (running)

- Nvidia GTX1070 for VM2 (running)

 

what doesnt work, using a gpu from a running VM for anything else, not for docker(s), not for another VM.

u can of course make 2 VM's using the same GPU, but NOT at the same time, so either VM1 OR VM2 using it ... or Docker(s) using it

(VM wont boot while assigned gpu is in use by another VM or Dockers, Dockers wont use hw transcode when gpu is in use by VM)

Edited by alturismo
more infos
Link to comment

Ok @alturismo, It seems like I will have to create a single Windows VM and run both applications that will need the single GPU. I dont have space for a separate GPU on my Mobo, just a single GPU is all I can do. I was thinking P2000 for both Plex and Blue Iris. 


If anyone has dealt with a similar scenario, please share your experiences. 


Thank you

Link to comment
  • 1 month later...

Very big grain of salt because I am searching the forums for answers to my GPU passthrough to my plex docker, BUT, from what I have read, dockers can share GPU's and be used concurrently, only limited by the session limit of the GPU, and for you and that lovely P2000, that's not an issue.

If I am wrong, somebody set me straight.

Link to comment
  • 6 months later...
22 minutes ago, Riverfrome said:

Hello Gents,

 

Sorry to come in again on this but i still was not sure...

 

My use case is i have a Nvidia P2000 assigned to plex but i now would like to ALSO use Emby... can i assign both plex and emby on the same GPU? and they would both be able to trans-code at the same time?>

 

Thanks

2 posts upper yours, yes ... 1 gpu for docker(s), 1 gpu for each VM (so each vm needs its own gpu on passthrough and cant be assigned to dockers or another vm)

  • Thanks 1
Link to comment

Thanks man... Like it was posted before, it's hard to find the answer... And sometimes it's good to just try stuff out on unraid and other times it's good to ask that stupid question.... "There is no such thing as stupid questions, unless you ask it twice"

Confirmed 1 GPU can be used for multiple dockers.

Link to comment
  • 1 year later...
  • 3 months later...

blueiris runs fine on CPU. been doing it for years.
i run BI in a windows 10 VM on unraid using 6 cores of my unraid server, it could run on 4 cores but i have a threadripper

I used to run BI on a weaker system too.

my 6 core VM runs 6 cameras. and hums below 30% cpu usage.

 

my GPU is for emby. and LOVE it.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.