Dynamic placement of virtual machines for cost optimization in multi-cloud environments

Abstract

The number of providers in the cloud computing market is increasing at rapid pace. At the same time, we are observing a fragmentation of the market in terms of pricing schemes, virtual machine offers and value-add features. In the early phase of cloud adoption, the price model was dominated by fixed prices. However, cloud market trend shows that dynamic pricing schemes utilization is being increased. In this plan, prices change according to demand in each cloud provider. In general, it is difficult for users to search cloud prices and decide where to put their resources. In this paper, we propose a scheduling model for optimizing virtual cluster placements across available cloud offers. This scheduler uses some variables such as average prices or cloud prices trends for suggesting an optimal deployment. Also, this scheduler is part of a cloud broker which automates actions and makes them transparent for users. The performance of our model is evaluated in a real-world cloud environment and the results show that user's investment decreases when part of the virtual infrastructure is dynamically distributed among clouds instead of maintaining it in a fixed one.