This duplicate data exists so that separate servers can be accessed by separate regions so that there’s no “bottlenecking” where everyone’s trying to access the same central server at once.