How severe does this issue affect your experience of using Ray?
- Medium: It contributes to significant difficulty to complete my task, but I can work around it.
https://docs.ray.io/en/master/cluster/kubernetes/user-guides/config.html#resources
It says to size the ray pod so that it take up the entire kubernetes node.
How should I size the worker processes in the Ray Pod ?
Should I create a process such that it takes up the entire Ray Worker Pod ?
Or is it safe to have multiple worker processes in a single Ray Worker Pod ?
If the Ray Pod size (logical) is of 8 CPUs. Can I write my 4 worker processes to take 2 CPUs each ?
scaling_config = ScalingConfig(
num_workers=4,
use_gpu=False,
resources_per_worker={"CPU": 2}
)