A multi-objective decision support framework for virtual machine placement in cloud data centers: a real case study

Abstract

Cloud computing (CC) is the concept of accessing to computing resources: servers, networks, storage, and applications, on demand through a network. This new paradigm has led to the birth of several data centers worldwide offering cloud services across millions of virtual machines. In fact, virtual machine placement (VMP) is considered as one of the greatest challenges for cloud providers to optimize their platforms in terms of physical machines number which reduces power costs and resources wastage. In this work, we propose an efficient framework based on multi-objective genetic algorithm (GA) and Bernoulli simulation that aims to minimize simultaneously used hosts and resource wastage in each PM on a CC platform. We operationalized our GA in a real case study related to the real cloud platform of the Office of the Merchant Marine and Ports of Tunisia (OMMP). This framework not only helped this company to optimize the VMP of their outsourced backup site, but also to minimize the operating expenses dedicated to the target platform. The proposed algorithm is tested on the OMMP’s data center, and experimental results show that the proposed technique significantly outperforms the compared methods especially in terms of VMP quality.