My environment is currently considering moving to a Clustering solution but I have a question I am unable to answer to at the moment. Our environment will be implementing DB's across the world (US, Hong Kong, UK, etc). Essentially what happens is in an application, a record is written to the DB, the application immediately ends, and then that new record is searched for in another application. Obviously with this much distance between servers there will be replication delays. So my question is, if we point the DB write at a Cluster IP, and attempt to look it up immediately via the Cluster IP, is it intelligent enough to know where to find this record? Or does it still need time for replication? For the sake of redundancy, if we write to Server A via Cluster IP, will it look up the record via Cluster IP in Server A every time? Or will it sometimes attempt Server B or C? Thanks for the assistance.