Efficient processing of OLAP queries using materialized views

Abstract

Data warehouses integrate information from multiple operational databases and are often used to support On-Line Analytical Processing (OLAP). OLAP queries involve a lot of aggregations on a large amount of data in data warehouses. A commonly used approach for efficient processing of OLAP queries is to store the results of frequently issued queries in materialized views and to make use of them in evaluating other queries. This approach involves selecting views to materialize and rewriting queries using materialized views. In this dissertation, we propose a new method to rewrite a given OLAP query using various kinds of materialized views that are available in data warehouses. We also propose dynamic management schemes for materialized views based on the past and future usability of the materialized views. We first define the canonical forms of OLAP queries and materialized views based on the selection and aggregation granularities, which are derived from the lattice of dimension hierarchies. Conditions for usability of materialized views in rewriting a given query are specified by relationships between the components of their canonical forms. We present a rewriting algorithm for OLAP queries that can effectively utilize materialized views having different selection granularities, selection regions, and aggregation granularities together. We propose algorithms to find an optimal and a near-optimal set of materialized views that results in a rewritten query which can be executed efficiently. We show the effectiveness and performance of the algorithms experimentally. We then propose caching schemes for materialized views in OLAP systems based on the usability of the query results in rewriting and processing other queries. For effective admission and replacement of materialized views, we consider the benefit of query results not only for recently issued queries but also for the expected future queries of the current query. We investigate the interactive and navigati...