To deal with the high resolution latencies in current existing mapping system, a hierarchical mapping system is proposed based on active degree. In the system, the mappings between the identifiers and locators are divided into three levels: active level, neutral level, and constant level. Based on these, a three tiers system architecture for mapping entries storing and resolving is designed. Stored mapping entries in different levels vary with the different active degrees of the remote communication terminal,and flow from one level to another. In order to minimize the mapping resolution latency, the construction model is proposed, which models the system construction process as a Markov Decision Process (MDP). Moreover, a Markov decision construction algorithm is proposed, which improves reinforcement learning to get the global optimal or near-optimal construction strategy. The simulation results show that the system has low resolve latency and good adaptability for network topology dynamic changes.%针对当前映射解析系统存在的映射解析时延过高的问题,该文依据终端的活跃程度,提出一种基于活跃度的分级映射解析系统.该系统将通信对端的身份位置映射信息划分为活跃级、中性级和稳定级3个等级,并据此建立了一种3层的映射解析存储架构,映射副本可根据自身活跃度的变化在3层之间动态调整存储位置.为最小化映射解析时延,在系统构建过程中,针对传统DHT构建方式存在的非位置感知问题,将系统构建过程建模为马尔科夫决策过程,并提出一种马尔科夫决策构建算法用于求解该模型.仿真结果表明,该系统能够显著降低映射解析时延和提升路由性能,对网络结构的动态变化具有良好的适应性.
展开▼
