POUR is a framework that accepts periodic information updates, collects information on demand, and accepts user-specified information while presenting a single unified view to the user. The primary functionality of POUR is through its query interface. A query consists of any number of XPaths, where each XPath returns a list of XML strings that satisfies the query. Depending on the XPaths specified and the contents of the POUR database, query processing may be as simple as a database lookup, or as complex as a series of queries down a POUR hierarchy to a set of POUR repositories that compute the requested information on the fly before the appropriate results are returned.

Users may use any valid XPath to retrieve results integrated from across the relevant periodic, on-demand, and user-specified sources. Periodic information comes from trusted sources at an unknown frequency such that any previous information from the same source can be completely overwritten. Ondemand information is computed when needed using queries to other POUR instances or using scripts dynamically executed by the Globus GRAM service. Finally, users may add their own information into a POUR repository if it conforms to a site-defined XML schema.

This program was written by Paul Kolano of Advanced Management Technology for Ames Research Center. For further information, click here or contact the Ames Technology Partnerships Division at (650) 604-2954.