I don’t think we as players will ever have the true “solution”.
The long term solution for CCP is to continue to do what they’ve been doing, to reorganise the backend into a microservices architecture that can scale horizontally across physical blades, and anything that isn’t directly gameplay, get’s moved off to other servers all together.
While I don’t know which hypervisor CCP are using, I can probably guess which one, and it (and all of the main competitors) now also provide container runtimes directly on the hypervisor with no server; and that’s probably where the code will end up also - microservices running in containers that can scale horizontally on a single blade and across blades where necessary.
That way it’ll be easier for CCP to scale the compute requirements needed to handle 10000 trying to get in system, with drones, cans and everything else that adds to server load.
It’s already so much better than it was even just a couple of years ago, but the solution is likely to be invisible to us, not just changes in mechanics.