Preferences

lhuser123 parent
And make it more complicated than K8s

jliptzin
Not possible
vajrabum
The platforms I've seen live on top of kubernetes so I'm afraid it is possible. nvidia-docker, all the cuda libraries and drivers, nccl, vllm,... Large scale distributed training and inference are complicated beasties and the orchestration for them is too.

This item has no comments currently.