In an XML Data Store (XDS), importing documents from external sources is a very frequent operation. Since a document import consists of a large number of individual node inserts, it is essentially a small bulkload operation. Hence, efficient bulkload support is crucial for XDSs. Essentially, XML bulkload is the transformation of an XML parser's output into the XDS's persistent storage structures. This involves two major subtasks: (1) Partitioning the documents' logical tree structure into subtrees smaller than a disk page in a way that is both space-efficient an suitable for later processing. (2) Mapping the subtrees to the XDS's internal page representation. In enterprise-scale environments with very large documents and/or very many parallel bulkloads, task (1) is particularly challenging, as not only disk space consumption, but also CPU and main-memory usage are important factors. In this article, we (1) discuss requirements for an XML bulkload module, (2) examine existing algorithms for tree partitioning with respect to their applicability as XML bulkload algorithms, (3) derive a new tree partitioning algorithm, (4) present the design and implementation of the bulkload module used in our Natix XDS, and (5) evaluate the implementation.
Additional information:
Das Dokument wird vom Publikationsserver der Universitätsbibliothek Mannheim bereitgestellt.
![[img]](https://madoc.bib.uni-mannheim.de/style/images/fileicons/application_pdf.png)
Search Authors in