Storage Service
A storage service defines the back-end implementation and storage characteristics of a SmartMapper ER Model.
A SmartMapper ER Model is only assigned to one storage service at a time. If the SmartMapper ER Model assigned to one service is assigned active to another service, it is automatically un-assigned from the first service and assigned to the new service.
JDBC-Based SmartMapper Service
The JDBC-based SmartMapper service allows you to store cross-referencing data in a database. Microsoft SQL Server, Oracle and MySQL are supported.
Currently, the driver of Microsoft SQL Server is packaged with BusinessWorks. If you want to use Oracle and MySQL, you need to add the drivers for them both at design time and runtime, see TIBCO ActiveMatrix BusinessWorks Plug-in for SmartMapper Installation for more information.
File-Based SmartMapper Service
The file-based SmartMapper service is very useful for prototyping and for production where the data is static. But you cannot write to the files, such as for seeding purposes, it is very slow. In addition, there is some risk of data corruption when doing concurrent writes.
While the JDBC-based service accesses a database, the file-based services do not. Instead, the file-based services are 100% cached. The file seeds the cache. Writes may be done for seeding and maintaining the file, but this is not a substitute for a database where there are large volumes of writes.
The file-based service runs inside the TIBCO ActiveMatrix BusinessWorks process engine and reads and writes the SmartMapper data in an XML file.
The file-based service does not perform well in write-intensive applications. The integrity of a file-based data store is lower than a database. File-based storage should not be used for data that cannot be regenerated.