Multi-objective materialized view selection using MOGA.

Materialized views are used as an alternative means for reducing the response time of analytical queries posed against a data warehouse. Since all views cannot be materialized and since optimal view selection is an NP-Hard problem, there is a need to select an appropriate subset of views for materialization that reduce the response times for analytical queries. This problem, referred to as view selection, is a widely studied problem in data warehousing. Several materialized view selection (MVS) algorithms exist that address the view selection problem, as a single objective optimization problem where the objective is to minimize the total cost of evaluating all the views (TVEC). This cost comprises two costs, i.e. the total cost of evaluation due to materialized views and the total cost of evaluation due to non-materialized views. Minimization of these two costs simultaneously would lead to the minimization of TVEC. In this paper, this bi-objective optimization problem, where the two costs are minimized simultaneously, has been solved using the Multi-Objective Genetic Algorithm (MOGA). The proposed MOGA based MVS algorithm selects the Top-K views from a multidimensional lattice with the purpose of achieving an optimal trade-off between the two aforementioned objectives. Materializing these selected Top-K views would reduce the response times for analytical queries and thereby would result in effective and efficient decision making. [ABSTRACT FROM AUTHOR]