Optimized Data Replication in Cloud Using Hybrid Optimization Approach

Abstract

Cloud computing (CC), in contrast to traditional high-performance computing environments, is a group of imaginary and networked resources of computing that are controlled by one unified maximum-performance computing power. Here, this work aims to develop a novel data replication method in the cloud. The data replication is carried out with a new multi-objective technique that considers constraints like cost, the distance between data centers, trust, and risk. Moreover, for optimal data replication, a new hybrid algorithm termed poor rich strategy assisted grasshopper optimization (PRS-GO) is introduced. To increase the accessibility of the system, the data used continuously should be duplicated in various areas. A minimal mean value of 0.66 is gained with the PRS-GO scheme, whereas Particle Swarm Optimization-Tabu Search (PSO + TS), Receding Horizon Control (RHC), Sun Flower Optimization (SFO), Cat Mouse-Based Optimization (CMBO), Hunger Games Search Optimization (HGSO), Seagull Optimization (SGO), Poor And Rich Optimization (PRO), and Grasshopper Optimization Algorithm (GOA) have got a high mean value of 0.722, 0.71, 0.71, 0.71, 0.7, 0.7, 0.7, and 0.69.