Exploiting the Structure of Update Fragments for Efficient XML Index Maintenance

XML databases provide index structures to accelerate queries on the content and structure of XML documents. As index structures must be consistent with the documents on which they are defined, updates on documents need to be propagated to affected index structures. This paper presents an index maintenance algorithm that is solely based on index definitions and update fragments instead of on the maintenance of auxiliary data structures. The use of index definitions assures that the algorithm supports arbitrary index structures defined on arbitrary document fragments. By exploiting the structure of update fragments, the algorithm directly extracts the nodes which are required for index maintenance from the fragments. Source queries are only necessary if the fragment does not contain all nodes required for indexing. The presented performance studies demonstrate the advantages of this approach over previous work that propagates each updated node individually.

Keywords:XML-Dokumente, Indexstrukturen, Updates