Energy-efficient Routing Simulation And Implementation In Large-scale IP Backbone Networks

With the development of broadband data services, the user numbers, bandwidth requirements and network traffic are ever-increasing. The energy consumption of network equipments has been steadily increasing. According to the investigation and study, the total power consumption of the world’s ICT (Information and Communication Technologies) equipment is growing steadily from168GW in2008to about430GW in2020. Thereinto the worldwide power consumption of network equipments is rapidly increasing at a yearly rate of12%. Meanwhile, the carbon emissions are also ever-rising with the growth of ICT industry. Currently, the carbon emissions of this industry account for about2%of the total global emissions, and this proportion is expected to double by2020. In addition, the problem of energy consumption is also one of the bottleneck problems faced by the current very large scale electric router. Therefore, the research on network energy efficiency has important significance.In the traditional design of network system, it enables the network withstand sudden peak load through the excess supply of resources, and improve network reliability through redundant links and equipments in response to the sudden fault failure. However, it doesn’t take the network energy efficiency into consideration. Hence, with the expansion of network size and the continuous update of network equipment, energy utilization of network has increasingly exposed some problems such as high energy consumption, low efficiency and high waste problems. According to the current different network framework, taking advantage of the energy-efficient network management and routing policy and under the premise of guaranteeing the quality of network performance and service to improve the network efficiency, then it can improve the utilization rate of network equipment and reduce the power consumption of standby equipment. This thesis has analyzed the energy efficiency of different network models, and puts forward the optimization model and routing strategy according to the relationship of power consumption and network traffic. First of all, we introduce the robust performance parameters of the network, the power model of network link the network energy efficient parameter and the energy consumption per bit. According to the energy efficiency of the IP network, the thesis makes it as the aim which is to minimize the network energy consumption per bit, makes a model efficiently, and proposes the energy-efficiency routing algorithm EEMCRA of load balance. The algorithm model mainly includes two parts:energy efficiency routing and load balancing. In order to further enhance network energy efficiency, we then put forward the energy-efficiency rerouting algorithm E2ReRA based on multi-restrain. By setting route weight using the power consumption model, we can get the maximum energy efficiency and guarantee the service quality of the network QoS through rerouting strategy.Secondly, according to multi-layer GMPLS network, we put forward the algorithm of topology control and routing strategy, and use sleeping algorithm method to conducts the power consumption of device port. In GMPLS network, according to the monitored network traffic and the router fault information, we can realize some links in the topology with no blocking of routing by using the calculating method of topology control and routing strategy in this algorithm. In this way, we can control node device power consumption so as to reduce network energy consumption. The extended LMP makes it possible to control the dormant of links.Finally, according to the IP over WDM network, by analyzing the power consumption of routers in IP layer and the power consumption of equipments in physical links of WDM layer, this thesis makes minimizing energy consumption per bit in the network as the optimization goal, and then establishes MILP optimization model. Through solving the optimization problems, it can obtain the minimum energy consumption per bit in the network so that improves the network energy efficiency. Because of the complexity of the IP over WDM network, this thesis uses auxiliary graph model to design the routing strategy with minimum network energy consumption. According to the business flow and network state, this strategy calculates the corresponding bit energy consumption cost matrix. And every times requests to establish LSP according to the smallest bit energy consumption.