So I’m heading down a path and I want to see if there’s already something out there before re-inventing the wheel.
What I’m looking to do is have 3 manager nodes that are “static” machines, as in full installed hosts with static IPs and such. I then want to have “dynamic” workers that can join and leave the swarm as needed. These “dynamic” nodes would be PXE booted and run as a LiveOS mode of Ubuntu Mini Remix.
The purpose is to have a watchdog process on the manager nodes watch the swarm and add nodes as needed to ensure running stacks have resources needed during times of demand, but then contract back down during periods of low demand by simply turning off the “dynamic” nodes and deleting them.
The “dynamic” nodes would just be VMs that are created and destroyed as needed.