All major relational database platforms have recently increased support for storing and managing XML natively. There are a few reasons for this beyond the marketing factor. An obvious one is the need for data to be commingled with XML in a manner that supports the relationships between the two. If this is true, then one might conclude that organizations are finding an increased need to customize the relationship between structured content and relational data. As a content management consultant I have found that not only is this statement true in the publishing industry it is becoming quite common. From the need to streamline publishing to multiple sources in multiple levels of granularity all the way to easing the editorial pain normally encountered when managing content and associated metadata.

How do I know all of this? I have worked recently on a number of projects that have taken advantage of this type of implementation. The basis for this case study is to pass along some lessons learned (both on the positive and negative sides) in one particular instance in which a system was architected using Microsoft SQL Server to house large XML documents modeled in a relational structure.

Due to the unique nature of the processes defined to support the assignment of metadata and the different levels at which this task is accomplished within the XML document model, we were forced to architect a custom solution to meet the clients needs.

This case study will take you through the analysis process and requirements that drove us to this solution as well as the technical architecture and implementation of the finished product. We will address everything from the DTD to the assembly of the XML for eventual output from the system. These and many other topics in between will guarantee a real world point of view on the inclusion of XML support within the RDBMS world and why it is such an important feature to embrace.