查看文章

Virtual Machine (VM) placement in a cloud data center is a Vector Bin-Packing (VBP) problem to minimize the number of PMs used for hosting the given VM requests. First-Fit-Decreasing (FFD) variants are widely used for VM placement. In this paper, a novel FFD variant, Aggregated Rank in FFD (FFD-AR) is proposed for VM placement. Simulation experiments were carried out using two datasets: a dataset inspired by Amazon EC2 instances and another is a synthetic dataset. The packing efficiency of the proposed FFD-AR results is better as compared to all the other baseline FFD variants. We believe the proposed FFD-AR can be applied to wide applications of VBP like production planning and logistics.