Materialised views aim to improve the response time of analytical queries posed on a data warehouse. This entails that they contain information that can provide answers to most of the future queries. The selection of such information is referred to as view selection. Several view selection algorithms exist in literature, most of which are greedy-based. In this paper, an answering query-based view selection approach (AQVSA), which considers both the size and the query frequency of each view, to greedily select top-k views for materialisation is presented. AQVSA first arrives at a reduced set of candidate views based on the query frequency of each view. This is followed by greedily selecting beneficial views from amongst these candidate views. Further, the experimental results show that AQVSA is able to achieve an acceptable trade-off between the total cost of evaluating all the views and the total number of queries answered by the selected views.
International Journal of Information and Decision Sciences – Inderscience Publishers
Published: Jan 1, 2013