Evaluating Contained Rewritings for XPath Queries on Materialized Views
Author(s)
Zhou, Rui
Liu, Chengfei
Li, Jianxin
Wang, Junhu
Liu, Jixue
Griffith University Author(s)
Year published
2011
Metadata
Show full item recordAbstract
In this paper, we study the problem how to efficiently evaluate a set of contained rewritings on materialized views. Previous works focused on how to find a set of contained rewritings given a view and a query, but did not address how to evaluate the rewritings on materialized views. To evaluate a potential exponential number of contained rewritings, we design two algorithms, a basic algorithm and an optimized algorithm. Both algorithms are built on the observation that the exponential number of contained rewritings are actually composed by a linear number of component patterns. In the optimized algorithm, we further ...
View more >In this paper, we study the problem how to efficiently evaluate a set of contained rewritings on materialized views. Previous works focused on how to find a set of contained rewritings given a view and a query, but did not address how to evaluate the rewritings on materialized views. To evaluate a potential exponential number of contained rewritings, we design two algorithms, a basic algorithm and an optimized algorithm. Both algorithms are built on the observation that the exponential number of contained rewritings are actually composed by a linear number of component patterns. In the optimized algorithm, we further design four important pruning rules and several heuristic rules that can effectively reduce the number of component patterns we need to evaluate. The experiments demonstrate the efficiency of our algorithms.
View less >
View more >In this paper, we study the problem how to efficiently evaluate a set of contained rewritings on materialized views. Previous works focused on how to find a set of contained rewritings given a view and a query, but did not address how to evaluate the rewritings on materialized views. To evaluate a potential exponential number of contained rewritings, we design two algorithms, a basic algorithm and an optimized algorithm. Both algorithms are built on the observation that the exponential number of contained rewritings are actually composed by a linear number of component patterns. In the optimized algorithm, we further design four important pruning rules and several heuristic rules that can effectively reduce the number of component patterns we need to evaluate. The experiments demonstrate the efficiency of our algorithms.
View less >
Journal Title
Lecture Notes In Computer Science
Volume
6587
Subject
Database systems