EPT: Energy Profit Threshold Metric and Algorithm for Energy Efficient Routing in SDN
share
Software defined networking (SDN) is rapidly growing networking paradigm in both industry and research areas, with network programmability as its powerful feature which enables propagating changes in the network easily. The logically centralized control in SDN enables dynamically minimizing the energy consumption of the links and the switches by diverting paths of packets. However, the links and switches are designed to accommodate maximum traffic volume and their power consumption is not traffic aware. Furthermore, energy efficiency and performance are opposite objectives that have to be addressed simultaneously. Addressing these issues, we propose an energy efficiency metric named Energy Profit Threshold (EPT) that considers utility of links, and provide IP formulation and heuristics for maximizing the EPT of a software defined network environment. We also analyze our approach considering different types of switches, namely virtual OvSwitch, NEC PF5240, and Zodiac FX. Experimental results show that maximizing the EPT value increases energy efficiency while maintaining acceptable network performance. The proposed heuristic algorithm achieves the highest EPT value and exhibits energy saving of more than 35 least power consumption, less average path length, and relatively better throughput as compared to similar utility based energy saving heuristics.
READ FULL TEXT