Do you have plans for multi-node persistence / geographic redundancy? We'd love something like this, but right now it would probably be easier to roll something

Do you have plans for multi-node persistence / geographic redundancy? We'd love something like this, but right now it would probably be easier to roll something custom on top of either Kafka or CockroachDB…