标题：An energy-aware ant colony optimization routing algorithm in the private network
作者：Kong, Guohong; Wang, Hua; Huang, Fuqiang; Yi, Shanwen; Wang, Yaqing
作者机构：[Kong, Guohong; Wang, Hua; Huang, Fuqiang; Yi, Shanwen; Wang, Yaqing] Shandong Univ, Sch Comp Sci & Technol, Jinan 250100, Shandong, Peoples R China.
会议名称：15th IEEE International Conference on High Performance Computing and Communications (HPCC) /11th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing (EUC)
会议日期：NOV 13-15, 2013
来源：2013 IEEE 15TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING AND COMMUNICATIONS & 2013 IEEE INTERNATIONAL CONFERENCE ON EMBEDDED AND UBIQUITOUS COMPUTING (HPCC_EUC)
关键词：Green network; power-aware; Ant cololny Optimization; intelligence; optimization
摘要：With the widespread of internet deployment and the increase of the users, energy consumption in wired network such as in the data center network, backbone computer network, financial network and so on, has become the main topic concerned in the industry and academic filed. However, energy efficiency is not an easy problem in wired network owing to the trade-off the network performance and save energy. Many researchers fall this problem in the class of multi-commodity flow problems which are NP-complete. To solve this problem, we propose a new method of using Ant Colony Optimization based on the graph growth traffic aggregation to find a solution containing all the communication sessions and minimize the number of nodes guaranteeing the bandwidth required by the source node, so as to power down some nodes in idle or light traffic period as many as possible based on the sleep-awake mechanism. The basic idea is to initialize randomly n ants. A solution path is kept in each ant, and the algorithm constantly evolves the solution to obtain a better result which minimizes the number of routing nodes and satisfies the quality of service (QoS) required by the users. Simulation results show that it is possible to reduce the number of nodes, and has better performance in searching and converging speed.