Use Repstance to distribute the load across several nodes, or to keep databases in several locations consistent.
How it works
- Repstance supports a Master-Master configuration that distributes the load across the nodes. Data is replicated between all nodes, so every application works with a local database.
- An additional node can be plugged into the configuration on the fly, which allows further expansion and load management while the other nodes keep running.
- Repstance supports a Multi-Master replication topology with the ability to ignore data produced by itself (loopback control), which ensures all instances are equal and any change made on one node is propagated to the other nodes.
- The nodes can be distributed across different geographical locations, so users work with a local database and have consistent databases in all locations. In the case of a failure, users can be switched quickly to any other available node, which holds all the changes replicated up to that moment.
Plan for conflicts
Repstance does not detect or resolve conflicts: if the same row is changed in two nodes at about the same time, the nodes can end up with different values (see Bi-Direction Replication). Plan the application so that a row is changed in one node at a time.
See also: Logical standby.