Indexing Temporal XML documents

Abstract

Different models have been proposed recently for representing temporal data, tracking historical information, and recovering the state of the document as of any given time, in XML documents. We address the problem of indexing temporal XML documents. In particular we show that by indexing continuous paths, i.e. paths that are valid continuously during a certain interval in a temporal XML graph, we can dramatically increase query performance. We describe in detail the indexing scheme, denoted TempIndex, and compare its performance against both a system based on a nontemporal path index, and one based on DOM.

Temporal XML: modeling, indexing, and query processing

“…This paper considerably updates and extends the work presented in [37]. Section 2 has been expanded, providing a better comparison of our model with other proposals.…”

“…In addition to the summaries themselves, we need indexes, access methods and additional data structures with information about the hierarchical relationships between nodes in a temporal XML documents. We present here TempIndex (introduced in previous work [37]), an indexing scheme that integrates LCP and interval summaries with additional indexes for efficient navigation.…”

“…In previous work [37] we addressed the problem of indexing temporal XML documents and introduced TempIndex, an indexing scheme for continuous paths that improves temporal query performance. In the second part of this paper we discuss TempIndex in detail.…”

“…Our proposal for a new class of summaries, which we call TSummary, adds the time dimension to the usual path summarization by considering continuous paths to element or value nodes. TSummary is the theoretical framework behind TempIndex, our indexing scheme for temporal XML data [37].…”

“…In the second part of the paper we present a framework for structural summaries of temporal XML documents, and study an indexing scheme, denoted TempIndex, introduced in previous work [37]. Several structural summaries have been proposed in order to optimize path query evaluation over non-temporal data graphs.…”