Human behavior-based particle swarm optimization for materialized view selection in data warehousing environment

Abstract

In a Data Warehouse setting, where space and repair costs are constrained by materialized view, Because it was nearly impossible to materialize all views, choosing the right MV was one of the smartest decisions in DW construction. At the same time, in the current world, methods for improving quilty of data warehouse, such as intelligence of the swarm, have appeared continually. As a result, this research proposes the first framework for reducing query response time using the algorithm of HPSO to determine the best view positions in the DW. As can be shown by comparing query response times on the data warehouse base tables to query response times on the MVs, the proposed method for choosing the best possible materialized views utilizing the HPSO outperformed all other algorithms. Base table queries take 14 times as long to implement as queries on MVs, according to this ratio. Queries sent via materialized viewes access take 106 milliseconds to respond, whereas those sent via direct access take 1066 milliseconds. This shows that queries accessed using MVs perform 1471.698 percent better than queries accessed via data warehouse.