University of Maryland
Source Selection for Wide Area Applications

Recent advances in technology has enabled wide area application (WAA) processing. This has caused a paradigm shift that has had a significant impact on the techniques used to select and locate information sources. In order to be scalable and flexible, WAA processing must be responsive to client specification of Quality of Access and Content (QoAC) requirements corresponding to the contents and capabilities of sources.

We consider a number of parameters of Quality of Access and Content (QoAC) metadata, to describe the contents and query processing capabilities of sources. Query processing at the client can be characterized by the latency (delay) to access the source, and other access costs associated with a source. Typical parameters for Quality of Access may include: TimeToFirst tuple; TimeToLast tuple; TimeToTopN tuple; $$ Access Cost; etc. Quality of Content metadata describe the source contents. They include the following: Domain information; Recency of Last Update and Frequency of Update; etc.

A characteristic of the wide area environment that poses a significant challenge to WAA processing is the wide variability in the availability of metadata. Sources are autonomous, and may not participate in a federation to provide such metadata. The dynamic WAN environment is typically characterized by wide variability of latency (delays) that depend on server and network workloads. Finally, rather than replica servers with identical contents, mirrors, portals and client caches differ considerably in their Quality of Content metadata.

Members of the Project

Recent papers

Please for recent unpublished papers on Source Selection for WAA.

A project on Efficient source selection and data access for users sharing limited bandwidth environments.

An overview of our research is in this KEYNOTE presentation ps or pdf presented at the 1999 Russian National Conference on Digital Libraries, St. Petersburg, October 1999.