Structural joins: a primitive for efficient XML query pattern matching

Al-Khalifa, Shurug; Jagadish, H. V.; Koudas, Nick; Patel, Jignesh M.; Srivastava, Divesh; Wu, Yuqing

doi:10.1109/icde.2002.994704

Cited by 474 publications

(532 citation statements)

References 15 publications

Supporting

Mentioning

528

Contrasting

Unclassified

Order By: Relevance

“…This system implements stack-based structural join algorithms [1]. It uses B+tree to implement the value index, and uses a hash indexing mechanism to implement the path indices.…”

Section: Methodsmentioning

confidence: 99%

“…This basic numbering scheme can be extended to include an additional number that encodes the level of the node in the XML data tree. This numbering scheme is used in our cost models when performing the structural joins [1,7,29] between parents and children or between ancestors and descendants. Figure 1 shows a data graph for a sample bib database using the Dietz numbering scheme.…”

Section: Models Of Xml Data Schema and Queriesmentioning

confidence: 99%

“…When XIST joins a path of two indexed subpaths, it uses a structural join algorithm [1], which guarantees that the worst case join cost is linear in the sum of sizes of the sorted input lists and the final result list. Let S be the set of indices which exclude the index on p 1 /p 2 , and C(p 1 /p 2 , S) be the cost of joining between path p 1 and path p 2 .…”

Section: Computing Evaluation Costsmentioning

confidence: 99%

“…The cost models used in this paper are developed for a prototype native XML DBMS that we are building. As in other native XML systems, this system stores XML data as individual nodes [6,12,15,18,29], and also uses stack-based join algorithms [1,2,7] for evaluating path expressions. However, the general framework of XIST can be adapted to systems with other cost models by modifying the cost equations that we use in this paper to accurately model the actual algorithms that are employed in that system.…”

Section: Introductionmentioning

confidence: 99%

“…A naive exhaustive traversal of the entire XML data to evaluate path expressions is expensive, particularly in large documents. Structural join algorithms [1,7,29] can improve the evaluation of path expressions, but as in the relational world, join evaluation consumes a large portion of the query evaluation time. Indices on XML data can provide a significant performance improvement for certain path expressions and predicates since they can directly select the nodes of interest.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

XIST: An XML Index Selection Tool

Runapongsa

Patel

Bordawekar

et al. 2004

Database and XML Technologies

View full text Add to dashboard Cite

Abstract. XML indices are essential for efficiently processing XML queries which typically have predicates on both structures and values. Since the number of all possible structural and value indices is large even for a small XML document with a simple structure, XML DBMSs must carefully choose which indices to build. In this paper, we propose a tool, called XIST, that can be used by an XML DBMS as an index selection tool. XIST exploits XML structural information, data statistics, and query workload to select the most beneficial indices. XIST employs a technique that organizes paths that are evaluated to the same result into equivalence classes and uses this concept to reduce the number of paths considered as candidates for indexing. XIST selects a set of candidate paths and evaluates the benefit of an index on each candidate path on the basis of performance gains for non-update queries and penalty for update queries. XIST also recognizes that an index on a path can influence the benefit of an index on another path and accounts for such index interactions. We present an experimental evaluation of XIST and current XML index selection techniques, and show that the indices selected by XIST result in greater overall improvements in query response times and often require less disk space.

show abstract

“…This system implements stack-based structural join algorithms [1]. It uses B+tree to implement the value index, and uses a hash indexing mechanism to implement the path indices.…”

Section: Methodsmentioning

confidence: 99%

Section: Models Of Xml Data Schema and Queriesmentioning

confidence: 99%

Section: Computing Evaluation Costsmentioning

confidence: 99%