A Multi-Weight Strategy for Container Consolidation

Abstract

Finding energy efficient management strategies while satisfying the Quality of Service (QoS) requirements is one of the critical challenges in Clouds. Virtual Machine consolidation has proved its efficiency in dealing with the energy issue in Cloud environments. Recently, containers are increasingly gaining popularity with regards to virtual machines (VMs) and going to be major deployment model especially at the edge of the network. Hence, it becomes critical to address energy efficiency from container’s point of view. Over the past few years, there have been several attempts for container consolidation heuristics. Unfortunately, most of the proposed strategies took container consolidation from VM’s point of view. In fact, they considered only one single deployment model in which containers are deployed on virtual machines. On the basis of such assumption, container consolidation is achieved through VMs consolidation. The fact stating that containers can be deployed as well in bare metal requires to take container consolidation from a different point of view. This paper focuses on container consolidation issue with regards to the bare metal deployment model and suggests a new consolidation strategy. The performance of our strategy is evaluated using the Docker Platform and is compared against two other consolidation strategies.