Fault Tolerance Aware Virtual Machine Scheduling Algorithm in Cloud Data Center Environment

Virtual machine (VM) scheduling is a complex problem in cloud environments, especially when reliability is considered. However, most of existing related works ignore the influence of this point. Therefore, this paper deeply explores fault tolerance aware VM scheduling (FTVS) and proposes an optimization model with multiple objectives and quality of service (QoS) constraints. Then, according to the characteristics of computing nodes (CNs) in CDC, a cost efficiency factor model of CN is defined and a heuristic algorithm based on best fit method is then proposed. Finally, to evaluate the efficiency and feasibility of the algorithm and models of this work, we use both simulation data sets and real-life cloud data center cluster data sets to conduct experiments. Experimental results show that the developed algorithm can improve successful execution rate of users’ VM requests and increase their overall satisfactions about cloud services.