Object-relational complex structures for XML storage

  • Authors:

  • Eric Pardede;J. Wenny Rahayu;David Taniar

  • Affiliations:

  • Department of Computer Science and Computer Engineering, La Trobe University, Bundoora VIC 3083, Australia;Department of Computer Science and Computer Engineering, La Trobe University, Bundoora VIC 3083, Australia;School of Business Systems, Monash University, Clayton VIC 3800 Australia

  • Venue:
  • Information and Software Technology
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

XML data can be stored in various database repositories, including Object-Relational Database (ORDB). Using an ORDB, we get the benefit of the relational maturity and the richness of Object-Oriented modeling, including various complex data types. These data types resemble the true nature of XML data and therefore, the conceptual semantic of XML data can be preserved. However, very often when the data is stored in an ORDB repository, they are treated as purely flat tables. Not only do we not fully utilize the facilities in current ORDB, but also we do not preserve the conceptual semantic of the XML data. In this paper, we propose novel methodologies to store XML data into new ORDB data structures, such as user-defined type, row type and collection type. Our methodology has preserved the conceptual relationship structure in the XML data, including aggregation, composition and association. For XML data retrieval, we also propose query classification based on the current SQL. Compared to the existing techniques, this work has several contributions. Firstly, it utilizes the newest features of ORDB for storing XML data. Secondly, it covers a full database design process, from the conceptual to the implementation phase. Finally, the proposed transformation methodologies maintain the conceptual semantics of the XML data by keeping the structure of the data in different ORDB complex structures.