A New Scalable and Energy Efficient Algorithm for VMs Reassignment in Cloud Data Centers

To improve resource utilization in Cloud Data Centers and in order to reduce energy consumption at the same time, reassignment of services is required and leads to efficient operational costs. This paper presents a new and scalable algorithm based on b-matching theory to judiciously replace resources (considered as Virtual Machines in our work) according to energy consumption constraints. Our algorithm is benchmarked by an exact approach based on an Integer Linear Program (ILP) formulation of the Bin-Packing problem. The b-Matching algorithm allows to find near-optimal solutions and scale for large problem instances.