You can use the new JDBC datasource to simplify the access of Java applications to sharded databases, where the applications can connect to the sharded databases transparently, with almost no change to the application code.
This feature enables Java connectivity to a sharded database without the need for a Java application to furnish a sharding key because the sharding data source derives the sharding key from the SQL statement. Your Java applications can scale out to sharded databases transparently as there is little to no change to the application code.
The sharding data source optimizes the performance of your applications as you do not need to configure the Universal Connection Pool (UCP) or create separate connection pools for cross-shard statements and single-shard statements. Also, you do not have to check-in or check-out a physical connection for every new sharding key because the sharding data source does it automatically.