Dynamic Server Switching for Energy Efficient Mobile Edge Networks
Qiuyun Wang, Qingyuan Xie, Nuo Yu, Hejiao Huang, Xiaohua Jia · 2019
Edge servers are densely deployed in the future mobile edge networks to meet the rapid increasing demand of mobile users. Since the distribution and traffic demand of user equipment (UE) fluctuate in time and over space, a number of edge servers may be underutilized which causes a great deal of energy waste. Therefore, we intend to reduce the energy cost of mobile edge networks, by dynamically switching on/off edge servers according to the variation of UEs' distribution. We formulate the energy saving problem in mobile edge networks as the minimum energy consumption (MinEn) problem which involves two critical issues: (1) cooperative service caching and UE association of adjacent BSs; (2) switching on/off edge servers. To solve the MinEn problem, we propose a dynamic server switching algorithm along with a lightweight UE distribution prediction mechanism. Simulation results show that our algorithm can greatly reduce the energy consumption of mobile edge networks compared with existing methods.