solidDB Help : solidDB Grid : Working with grids : Scaling a grid
  
Scaling a grid
Note Check the solidDB Release Notes for any limitations that are associated with using a grid in the current release.
While a grid consists of only one node, the node behaves just like a standalone solidDB server. Tables can be partitioned or replicated but those declarations have no effect.
When one or more nodes are added to a grid, the following actions occur automatically:
For a partitioned table, primary partitions are assigned to the new nodes to spread the write load.
For a non-partitioned (shared) table, the entire table is replicated to the new nodes in order to provide high availability in case of failure.
Client drivers for ODBC and JDBC (‘connectors’) detect the new nodes and start to route queries to them.
Note If you use the standard connection string format a driver connects to the specified node. If you use the grid-specific connection string format, the driver can select the node based on the operation that is being executed.
See
Adding a node to a grid
Removing a node from a grid
Disconnecting a node from the grid temporarily
Reconnecting a node to the grid
Go up to
Working with grids