Special application requirements
All of the solutions covered so far are master/slave solutions that, while asynchronous, they expect fairly tight contact between all the database nodes. For example, the master in a Hot Standby deployment will eventually run out of disk space and crash if it can't ship WAL files to the slaves it's feeding.
Bucardo
One of the things most PostgreSQL replication solutions have in common is that they have a single master node. Bucardo is instead a multi-master replication solution, also implemented with triggers; refer to http://bucardo.org/.
Anytime you have more than one master, this introduces a new form of replication problem. What do you do if more than one node modifies the same record? Solving that is called conflict ...
Get PostgreSQL 9.0 High Performance now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
