Replication Based Fault Tolerance Approach for Cloud

Abstract

The usage of cloud computing has been increasing everyday and in almost every organization. It is now an integral part of the information technology market. And this rapidly growing need gives rise to consider two main criteria namely reliability and availability to sustain growing demand. To ensure SLAs, and gain customer confidence, cloud architectures use fault tolerant strategies. The work in the literature towards fault tolerant cloud has not considered aspects such as using resource subsets to be fault tolerant. In this paper, we have used a replication based fault-tolerance strategy which considers cost and fault proneness of VMs. Simulation of the proposed algorithm has been carried out using CloudSim and the result shows an improvement in average total execution time and average delay time.