Query Plan Generation in DDS Using Non Dominant Based Teacher-Learner Optimization (ND-TLBO) Algorithm

摘要

The query optimization issue in large-scale distributed datasets are NP and difficult to resolve. The demand for the optimizer upsurges as the number of relations and positions in a distributed database query processing rises. Investigation are performed so as to obtain a suitable methodology to pursue an optimum output particularly when the dimension of the dataset upsurges. The foremost issue for the need of query optimization in any disseminated datasets systems is to diminish the total query processing cost for the specified queries. This study, proposed a novel approach known as Non Dominant Based Teacher-Learner Optimization algorithm (ND-TLBO) for the multi objective test functions in the distributed query processing as to obtain the optimized query plans. TLBO is a parametric less optimization algorithm which is enhanced in such a way that in order to employ the multi objective test function into the approach, the dominance property is used between the individuals. Fast non dominated ranking algorithm and crowding measure are the two dominance property employed in this approach. The performance of the suggested methodology is matched with the prevailing TLBO query plan generation technique and multi objective GA based query plan generation technique.