Energy efficient container consolidation method in cloud environment based on heuristic algorithm

Abstract

While container-based clouds are increasingly gaining popularity, minimizing the power consumption of the data center is still one of the major challenges for cloud providers. Dynamic consolidation of containers presents a significant opportunity to save energy in data centers and there are researches use optimization metaheuristic algorithm to find a near-optimal solution for this problem. However, the computation time and complexity of such algorithms increase exponentially with the number of containers. In this paper, a heuristic dynamic container consolidation method which consists of four algorithms that solve the problems in different stages of container consolidation has been proposed. We migrate redundant containers from the hosts before they overload and place them to other host to guarantee QoS requirements. An adaptive reserved resources to prevent re-overload of hosts has also been applied. Experimental results demonstrate that our proposed approach can lead to further energy saving with QoS guarantees compared with some existing approaches.