Affordable Access

Navigierbare Kompression von XML-Datenströmen

Publication Date


document01.dvi Navigierbare Kompression von XML-Datenströmen Dissertation zur Erlangung des Grades eines Doktors der Naturwissenschaften der Fakultät für Elektrotechnik, Informatik und Mathematik der Universität Paderborn vorgelegt von Rita Hartel Paderborn, September 2008 Abstract Nowadays, XML has shown to be the de facto standard for electronic data interchange on the Internet. Available XML data ranges from small Web pages to possibly unbounded streams, as used e.g. in news agencies. Especially when using small mobile devices (such as mobile phones or PDAs), the data size forms a problem due to the limitations in main memory, and the size of the transferred data forms a problem due to the limitations in energy consumption. In these cases, i.e., whenever the data size or energy consumption limitations form the bottleneck of an XML based application, these applications can pro�t from the usage of XML compression. It is desirable that these applications can perform all XML based operations, like XML query evaluation and XML data manipulation, directly on the compressed XML data, to avoid additional computation caused by prior decompression and subsequent compression. Furthermore, it is desirable that there is no loss in e�ciency when performing query evaluation on compressed data in contrast to performing query evalution on uncompressed data. Finally, it is desirable that it is possible to perform the query evaluation on possibly unbounded streams. The existing approaches to XML compression can be classi�ed according to whether they support these XML based operations or not. Approaches like GZip, BZip2, XMill, and others reach a strong compression, while the eva- luation of queries requires prior decompression and subsequent compression. Other approaches, like XGrind and XQueC, allow to evaluate queries on the compressed data, but they are outperformed in terms of compression ratio by those approaches that do not support these X

There are no comments yet on this publication. Be the first to share your thoughts.


Seen <100 times