Joint Placement of Gateways and Controllers in SDN-Enabled Space-Ground Integration Network
Yizhou Shen, Wen Chen, Jiawei Liu · 2022
The integration research of terrestrial and satellite network utilizes Software-Defined Networking (SDN) to build an emerging network architecture. Therein, the joint placement problem of gateways and controllers is crucial for the design of this SDN-enabled integrated network. Different from the previous work which only considers the network latency and reliability, we introduce the load average difference rate metric to balance the load in the network. In order to minimize the load average difference rate and determine the appropriate number of controllers and gateways, we propose a simulated annealing algorithm based on improved density peak clustering (DBSAA). We conduct experiments on real-world network topology to verify the effectiveness of our algorithm for the joint placement problems. The simulation results show that compared with the Optimal Enumeration Algorithm (OEA) and Simulated Annealing Algorithm (SAA), DBSAA can obtain approximately optimal performance in a shorter running time.