UDTT by nature supports session failover in distributed clustered servers or cloud environments. When the UDTT property sessionPersistence is set to true in btt.xml, the UDTT session context and flow control information will be stored in the HTTP session as a SessionEntry attribute. When an HTTP session is cached by failover support of the Application Server, UDTT then also supports failover, based on the platform of Application Server. In the past, the IBM WAS ND had this feature, and now in cloud environment, IBM Websphere Liberty can also leverage the third-part cache services to achieve this. For more information, please refer to the document:

When the platform stores the HTTP session objects in cache, the java Serialization / Deserialization APIs is called. Any data objects which are required to be persisted are required to implement the java Serialization/Deserialization interface methods.