A multi-objective scheduling strategy based on MOGA in cloud computing environment

Cloud service providers usually utilize the resource in the data center to offer cloud computing service which needs to satisfy the demand of different customers. The load among the physical servers in the data center needs to be balanced to avoid hotspot and improve resource utility. In this paper we present a load balancing system and design an algorithm base on MOGA to get a new mapping relationship between physical machines and VMs. By a set of migrating operations of VMs the problem of load imbalance are solved. Through comprehensive simulations, the experimental results demonstrate that our proposed approaches can significantly improve the resource utilization when system load is stable variant.