Hi, I'm trying to figure out a "not quite prod, but better than dev" deploy structure. I have a couple of questions: 1. It looks like the scheduler and server use the most memory. Does it make sense to put those on beefier machines? I assume that they can not be duplicated so one machine for each (or both on one). 2. Is there a way to get the workers to start on a different machine? In other words, is there a worker pool I can start or is there a way to use bin-packing to get the most out of the worker machine?

we’re currently working on improving our kube deployment in production! So separating the workers on a different machine is definitely something that will be possible in the future

Thanks for the info. I’m actually working with Nomad. If more familiar with the Hashi stack and k8s is often overkill for my needs. I would imagine changes made for k8s would also translate to Nomad though