Hi,
for some time now I am running this project that uses GPU resources as a main basis.
I have several docker containers running, and each use different amount of GPU and VRAM.
Is there a way to monitor GPU usage of those containers each with zabbix?
f.e. container1 uses 18% of GPU and 2GB of VRAM, container2 uses 60% of GPU and 1GB of VRAM.
My Zabbix sees overall usage of GPU = 78% and 3 GB VRAM, but not separately for each container.
Is there a way?
The only thing I came up with would be installing separate agents inside those and adding those containers as different hosts, but didn't test it and don't know if it'd work.
Also what if there will be 1000 containers like this?
for some time now I am running this project that uses GPU resources as a main basis.
I have several docker containers running, and each use different amount of GPU and VRAM.
Is there a way to monitor GPU usage of those containers each with zabbix?
f.e. container1 uses 18% of GPU and 2GB of VRAM, container2 uses 60% of GPU and 1GB of VRAM.
My Zabbix sees overall usage of GPU = 78% and 3 GB VRAM, but not separately for each container.
Is there a way?
The only thing I came up with would be installing separate agents inside those and adding those containers as different hosts, but didn't test it and don't know if it'd work.
Also what if there will be 1000 containers like this?