SPAX: Simple Path based XML Data Storage and XPath Evaluation
Authors
- Changhao Liu (IBM, China)
- Shuo Wu (IBM, China)
- Guogen Zhang (IBM Silicon Valley Lab, USA)
- Guoming He (Renmin University of China, China)
Abstract
An XPath-based XML query is typically applied on multiple XML documents, which are stored in an XML repository either in a native XML database or in an XML-typed column of a table in a relational database. Such XML queries are usually used to access pieces of information, which are small in size and scatteredly stored in most cases, from the entire XML documents. Taking this characteristic into consideration, this paper presents an XML data management system, called SPAX, which adopts a Simple Path clustering storage solution. Accordingly, a novel XPath evaluation approach is introduced. The system can avoid retrieving unneeded data into memory, reduce I/O times, and thereby enhance the XPath evaluation. Extensive experimental results reported in this paper demonstrate that the approach is promising and can achieve significant performance improvements.
Session
EDBT Industrial Session 1: Industrial Session (Wednesday, March 25, 16:00—17:30)