toward an enhanced dynamic vm consolidation approach for cloud datacenters using continuous time markov chain

Dynamic virtual machine (vm) consolidation is an effective manner to reduce energy consumption and balance the resource load of physical machines (pms) in cloud data centers that guarantees efficient power consumption while maintaining quality of service requirements. reducing the number of active pms using vm live migration leads to prevent inefficient usage of resources. however, high frequency of vm consolidation has the negative effect on the system reliability and we need to deal with the trade-off between energy consumption and system reliability. in recent years many research work has been done to optimize energy management using power management techniques. although these methods are very efficient from the point of view of energy management, but they ignore the negative impact on the system reliability. in this paper a novel approach is proposed to achieve a reliable vm consolidation method. in this way, a markov chain model is designed to determine the reliability of pms and then it has been prioritized pms based on their cpu utilization level and reliability status. two algorithms are presented to determining source and destination servers. the efficiency of our proposed approach is validated by conducting extensive simulations. the results of the evaluation clearly show that the proposed approach significantly improve energy consumption while avoiding the inefficient vm migrations.