Instance-Independent View Serializability for Semistructured Databases

TL;DR

This paper introduces a document-independent approach to concurrency control in semistructured databases, enabling efficient verification of schedule equivalence and improving transaction management for hierarchical data like XML.

Contribution

It presents a novel, polynomial-time decidable method for view serializability that is independent of specific document structures in semistructured databases.

Findings

Polynomial-time decidability of schedule equivalence

Efficient view serializability verification for semistructured data

Applicable to hierarchical data like XML documents

Abstract

Semistructured databases require tailor-made concurrency control mechanisms since traditional solutions for the relational model have been shown to be inadequate. Such mechanisms need to take full advantage of the hierarchical structure of semistructured data, for instance allowing concurrent updates of subtrees of, or even individual elements in, XML documents. We present an approach for concurrency control which is document-independent in the sense that two schedules of semistructured transactions are considered equivalent if they are equivalent on all possible documents. We prove that it is decidable in polynomial time whether two given schedules in this framework are equivalent. This also solves the view serializability for semistructured schedules polynomially in the size of the schedule and exponentially in the number of transactions.