Multi‐objective materialized view selection using flamingo search optimization algorithm. (2nd December 2022)

• Record Type: Journal Article
• Authors: Srinivasarao, Popuri; Satish, Aravapalli Rama
• Is Part Of: Software, practice & experience. Volume 53:Number 4(2023)
• Page Start: 988
• Page End: 1012
• Publication Date: 2022-12-02
• Subjects: data warehousing -- flamingo search algorithm -- materialized view selection -- query processing; Computer software -- Periodicals; Computer programming -- Periodicals; Computer programs -- Periodicals; 005.3
• DOI: 10.1002/spe.3174
• Languages: English
• ISSNs: 0038-0644

Abstract

Abstract: A materialized view selection in data warehouse management is important to speed up query processing. The data presented in data warehouses is generally stored as a set of materialized views. The major challenge is determining which views to materialize and satisfy the response time with reduced cost functions. This paper proposed an effective multi objective cost model based flamingo search algorithm for materialized view selection in data warehouse design. The multiple view processing plan structure of the data warehouse describes the search space of problem in order to select the optimal materialized views. The proposed model evaluates a multi‐objective optimization problem based on the cost functions resulting from materialization. The multiple objective functions of the proposed model are maintenance costs, current query processing costs, response cost and previous query processing costs. This model selects the top‐k views for materialization by satisfying the mentioned multi‐objective functions. The experimental results are simulated using the TPC‐H dataset. The efficacy of proposed model is measured by comparing the obtained results of the proposed model with various existing approaches.