An efficient co-operative framework for multi-query processing over compressed XML data

XML is a de-facto standard for exchanging and presenting information on the Web. However, XML data is also recognized as verbose since it heavily inflates the size of the data due to the repeated tags and structures. The data verbosity problem gives rise to many challenges of conventional distributed database technologies. In this paper, we study the XML dissemination problem over the Internet, where the speed of information delivery can be rather slow in a server-client architecture which consists of a large number of geographically spanned users who access a large amount of correlated XML information. The problem becomes more severe when the users access closely related XML fragments, and in this case the usage of bandwidth is inefficient. In order to save bandwidth and process the queries efficiently, we propose an architecture that incorporates XML compression techniques and exploits the results of XPath containment. Within our framework, we demonstrate that the loading of the server is reduced, the network bandwidth can be more efficiently used and, consequently, all clients as a whole can benefit due to savings of various costs.