EFFICIENT RESOURCE ALLOCATION IN CLOUD DATA CENTER USING HYBRID MULTI-OBJECTIVE OPTIMIZATION ALGORITHMS

Abstract

The amount of data that is generated daily is incredibly huge, and this leads to high demand on cloud storage, hence, many cloud data centers. Using electricity at this massive scale in these data centers has made the operations very expensive. In both, the aim is to reduce the energy used so that it does not surpass the SLA and lead to a breach of the same. Virtual machine placement (VMP) is the method that labels the approach that ought to be followed to determine which virtual machine will be positioned in which host of the data center to be executed. The current methods rely on the energy consumption, whereas the breaches of SLA on the Virtual Machine Placement are not taken into consideration. The challenge in this paper is to consider the distribution of resources at cloud data centers through VM placement by means of a multi-objective optimization algorithm. To address the issue with the combination of two algorithms, Non-Dominated Sorting Genetic Algorithms II (NSGA-II) and Multi-Objective Particle Swarm Optimization (MOPSO), in order to come out with the hybrid variant, that is, Hybrid Multi-Objective Optimization Algorithms (HMOOA). The outcomes of the experiments prove that the proposed algorithm is superior in the aspects of the maximum energy efficiency and the minimum rate of SLA Violation as compared to the existing one. It equally devises methods of reducing the movement of VMs as it allocates VMs.