Scalable XSLT Evaluation

TL;DR

This paper introduces a Streaming Processing Model (SPM) for XSLT evaluation, enabling efficient, memory-friendly processing of large XML documents and results, significantly outperforming existing methods.

Contribution

The paper proposes a novel SPM approach for XSLT processing that reduces memory usage and improves performance on large XML data sets.

Findings

SPM improves XSLT evaluation speed 2 to 10 times.

SPM enables processing of large XML documents without extra memory buffers.

Experimental results confirm high scalability and efficiency of SPM.

Abstract

XSLT is an increasingly popular language for processing XML data. It is widely supported by application platform software. However, little optimization effort has been made inside the current XSLT processing engines. Evaluating a very simple XSLT program on a large XML document with a simple schema may result in extensive usage of memory. In this paper, we present a novel notion of \emph{Streaming Processing Model} (\emph{SPM}) to evaluate a subset of XSLT programs on XML documents, especially large ones. With SPM, an XSLT processor can transform an XML source document to other formats without extra memory buffers required. Therefore, our approach can not only tackle large source documents, but also produce large results. We demonstrate with a performance study the advantages of the SPM approach. Experimental results clearly confirm that SPM improves XSLT evaluation typically 2 to 10 times better than the existing approaches. Moreover, the SPM approach also features high scalability.