Configuring distributed DataServer applications using ProBroker

The figure below shows an example of a distributed DataServer configuration that uses ProBroker. The figure illustrates how to combine clients and servers that run on different platforms.

In the figure above, the OpenEdge client runs on a client machine (either in Windows or on UNIX) and can access a remote Windows server. Multiple brokers can be running simultaneously on one server machine. The client connects initially to the ProBroker executable by locating the host and port of the executing broker. It is then automatically reconnected to a DataServer established for it by the broker. Each executing broker can spawn a multitude of DataServer processes. A spawned DataServer process uses the ODBC driver to locate and connect to the data source. The actual target database might be either local or remote to the host machine. Note that in this example, the schema holder also runs on the Windows client; however, you can optionally locate it on any host that is accessible on your network. Each spawned DataServer can service database requests for the same database or for a different database than those of other spawned servers.

In remote DataServer connections, there is always a one to one relationship between a specific DataServer client and its DataServer server process. OpenEdge handles the communication between the client and the server. The client and server processes that make up the DataServer adapt to a variety of network configurations.