Distributed Search

Distributed Search allows search engines to operate in two or more separate processes, distributing the workload across either multiple CPUs on a single server machine or on multiple server machines. Distributed Search is most practical for very large datalists or when search engines need to be started and stopped separately from the main application.

Distributed Search is best suited for loading large datalists comprising multiple millions of names. Distributed Search engines each manage a part of a single large datalist.

If you need to load multiple datalists into a single server application, Embedded Search may be a more effective use of computer resources.

Applications can communicate with Distributed Search search processes either directly through TCP/IP connections or through applications built with IBM® NameWorks.