Energy-Efficient Manycast Routing and Spectrum Assignment in Elastic Optical Networks for Cloud Computing Environment

In this paper, we present an energy-efficient manycast routing and spectrum assignment (EEM-RSA) algorithm in elastic optical networks supporting cloud computing applications. The proposed EEM-RSA is adapted for both static and dynamic scenarios. First, an integer linear programing formulation is derived for energy-efficient manycasting and spectrum assignment; then, the corresponding heuristic methods are proposed. To reduce the energy consumption, inactive (idle) elements are turned off, and in the proposed energy-efficient manycasting heuristic, the number of activated elements are minimized. The power consumption of network elements is modeled by considering a constant overhead for the element activation and a variable traffic-dependent term for the element operation. Furthermore, two types of data centers based on their power supply, renewable (green) or nonrenewable (brown), are considered to explore the capability of the proposed algorithms in decreasing green house gases emission. To investigate the impact of renewable energy sources, two approaches are utilized in the destination selection (DS) phase of the EEM-RSA, namely green-energy aware-DS and green-energy unaware-DS. All heuristic algorithms are evaluated by using an event-driven simulator based on the Poisson traffic model. The simulation results reveal that by applying the proposed energy-aware heuristic algorithm, the network energy consumption is reduced at the cost of increasing the blocking probability. However, by designating the shortest path, instead of the path with the lowest power consumption, the blocking probability is reduced, whereas increasing the energy consumption. Thus, we introduce an intermediate solution, referred to as blocking-aware energy-efficient manycasting, which compromises between the power consumption and blocking probability performance metrics.