i've always wondered : it seems to me that handling hordes of process is today made quite simple thanks to tools like kubernetes, and middlewares like queues or

i've always wondered : it seems to me that handling hordes of process is today made quite simple thanks to tools like kubernetes, and middlewares like queues or databases of various kinds for process 2 process communication with various guarantees. What's not solved by erlang is IMHO the hard part : how does your system ensures data persistence, availability and consistency ? Rolling my own cassandra or kafka or distributed db isn't really something that enchants me, and i don't see how the beam can makes this any easier ( they are fundamental problems, not technological ones).