Hybrid algorithm using the advantage of krill herd algorithm with opposition-based learning for dynamic resource allocation in cloud environment

Abstract

The cloud computing systems have more consideration due to the growing control for elevated concert computing and data storage. Resource allocation plays a vital role in cloud systems. To overcome the obscurity present in resources allocation system. In this paper, we design and develop a technique for dynamic resource allocation. A Hybridized approach is designed with the help of multi-objective oppositional krill herd optimization algorithm (OKHA). It is a combination of the krill herd algorithm and Opposition-based learning (OBL), OBL is added to get enhanced performance of the krill herd algorithm. The objective of this hybridization is to reduce the cost. In this Hybridized process each task consists of two cost i.e monetary cost and computational cost. Here each task is divided into many subtasks and assigns the respective resources to it. Our proposed multi-objective optimization algorithm will decide allocation of resource for the each subtask in this process. Finally, the testing is passed out, we evaluate our proposed algorithm with PSO, and GA algorithm we verified the performance levels of our proposed Multi-objective optimization algorithm.