Designing Schemaless, Uber Engineering's Scalable Datastore Using MySQLWhile all three systems are able to scale linearly by adding new nodes online, only a couple systems can also receive writes during failover. None of the solutions have a built-in way of notifying downstream dependencies of changes, so we would need to implement that at the application level. They all have indexes, but if you’re going to index many different values, the queries become slow, as the
