Energy Management in IP Traffic Engineering with Shortest Path Routing

Even if the Internet is commonly considered a formidable mean to reduce the impact of human activities on the environment, its energy consumption is rapidly becoming an issue due to the exponential traffic growth and the rapid expansion of communication infrastructures worldwide. One of the first options available to reduce energy consumption is that of limiting wastes. Actually, even if it has been shown that traffic load greatly varies over time and rarely saturates available capacity, the energy consumed by the network is almost constant as if it were fully loaded at all time. In this paper we propose an IP traffic engineering approach that allows to adapt the network energy consumption to different daily traffic scenarios switching on and off communication interfaces (links) and entire routers. We focus on routing domains of the Internet Service Providers where the popular OSPF (Open Shortest Path First) protocol is adopted and we consider the problem of switching off network elements (links and routers) and of adjusting link weights so as to minimize the energy consumption and the network congestion. We present two heuristics for this problem: the Greedy Algorithm for Energy Saving (GA-ES) and the Two-stage Algorithm for Energy Saving (TA-ES). The computational results for three real network topologies show that it is possible to switch off up to 80% of the core nodes during low traffic periods (night hours), and up to 65% during moderate traffic periods, while guaranteeing the same point-to-point service quality, and moderately increasing the network congestion.