A Distributed Fault Tolerant Algorithm for Load Balancing in Cloud Computing Environments

Abstract

Cloud computing is a promising paradigm that provides users with higher computing benefits in terms of cost, availability and flexibility. Nevertheless, with potentially thousands of connected machines, faults become more frequent and may have an adverse effect on the application. Consequently, fault-tolerant load balancing becomes necessary in order to optimize resource utilization while ensuring the reliability of the system. Different approaches have been proposed in the literature for fault tolerance in cloud computing. However, they suffer from several shortcomings: some fault tolerance techniques use task replication which reduce the cloud’s efficiency in terms of resource utilization. While other models rely on checkpoint recovery to tolerate failures, resulting in an increase in the mean response time. To address these shortcomings, an efficient and adaptive fault tolerant algorithm for load balancing is proposed. Based on the CloudSim simulator, some series of test-bed scenarios are considered to assess the behavior of the proposed algorithm.