Efficient Evaluation of NOT-Twig Queries with Parent-Child Edges in A Tree-Unaware RDBMS

Despite a large body of work on XML query processing in relational environment, systematic study of NOT-twig queries has received little attention in the literature. Such queries contain notpredicates and are useful for many real-world applications. In this paper, we present an efficient strategy to evaluate NOT-twig queries with parent-child edges in a schema-oblivious and tree-unaware relational environment. Devising efficient XML query processing strategy on “purely relational" environment is a challenging problem as recent results demonstrate that query performance often degrades due to the generation of inefficient query plans. Our proposed technique is built on top of the SUCXENT++ system, a tree-unaware approach designed primarily for read-mostly workloads. It extends the encoding scheme of SUCXENT++ by adding two new labels, namely AncestorValue and AncestorDeweyGroup, that enable us to efficiently filter out elements satisfying a not-predicate by comparing their ancestor group identifiers. In this approach, a set of elements under the same common ancestor at a specific level in the XML tree is assigned same ancestor group identifier. Based on this encoding scheme, we propose two variants of a novel SQL translation algorithm for NOT-twig query evaluation, namely HYBRID and PURE. Experiments carried out confirm that our proposed approach built on top of an off-theshelf commercial RDBMS significantly outperforms representative tree-unaware and native approaches. Interestingly, our scheme reduces significantly the performance gap between a representative tree-aware (MonetDB/XQuery) and tree-unaware approaches and, surprisingly, even outperform it for certain NOT-twig queries.