Today, the main reason for replication in distributed database systems is the improvement of system performance. Additionally, replication enhances the reliability and availability of a system. For this purpose it is often necessary to adapt an heterogeneous legacy system in a company network. Because of these preconditions synchronous replication is not the first choice, as it rarely supports heterogeneous systems and the complex algorithms make the systems slower rather than faster. The solutions for this problem are asynchronous replication algorithms. Therefore, asynchronous replication and its application in a real world company are at the center of this thesis. A practical realization shows an implementation of a new replication system for the Raiffeisenverband Salzburg. The replication system is built upon a customized message queuing middleware and a self implemented endpoint application. In the future it should replace the outdated existing system and allow an expansion of the existing replication system. At the moment the system is partly implemented and in the test stage.