The architecture of the Obelix-an improved Internet search engine

Contemporary Internet search engines can hardly cope with enormous amounts of indexed data, and very seldom succeed in extracting useful information. In order to anticipate the usefulness of a Web page, the proposed solution keeps track of users' actions on a Web page. A modified Web client informs a dedicated server of users' actions performed on a Web page, such as printing, bookmarking or mailing. The search algorithm relies on this information when ranking pages according to their usefulness. The paper describes the proposed paradigm in detail and focuses on the architecture of the server dedicated to collecting information of users' actions. The server's architecture is based on reconfigurable technology and VCC HOT II reconfigurable boards.