Materialized View Selection for Query Performance Enhancement Using Stochastic Ranking Based Cuckoo Search Algorithm

摘要

Materialized view selection (MVS) improves the query processing efficiency and performance for making decisions effectively in a data warehouse. This problem is NP-hard and constrained optimization problem which involves space and cost constraint. Various optimization algorithms have been proposed in literature for optimal selection of materialized views. Few works exist for handling the constraints in MVS. In this study, authors have proposed the Cuckoo Search Algorithm (CSA) for optimization and Stochastic Ranking (SR) for handling the constraints in solving the MVS problem. The motivation behind integrating CS with SR is that fewer parameters have to be fine tuned in CS algorithm than in genetic and Particle Swarm Optimization (PSO) algorithm and the ranking method of SR handles the constraints effectively. For proving the efficiency and performance of our proposed algorithm Stochastic Ranking based Cuckoo Search Algorithm for Materialized View Selection (SRCSAMVS), it has been compared with PSO, genetic algorithm and the constrained evolutionary optimization algorithm proposed by Yu et al. SRCSAMVS outperforms in terms of query processing cost and scalability of the problem.