Compression of XML Data
Mathias Neumüller · UPT. Syiah Kuala University Library (Syiah Kuala University) · 2001
Most implementation of the Document Object Model (DOM), which are used to represent XML data in the computer's memory, suffer from extensive memory requirements. Thus large volumes of data are hard to handle. This work tries to analyse the reasons for this and designs a system that is specifically tailored for the need of large document. A DOM compliant data structure called DDOM, that uses dictionary compression to reduce the amount of memory being used, was designed, implemented and tested. Comparison with other implementations show a saving in the order of 30% to over 80% in terms of memory usage. The influence of different types of data on the compression ratio is shown. This report also identifies the the conceptual differences between relational and irregular data sources with respect to the possibilities of a compact representation.