基于聚类的ICN缓存空间规划方法

ICN CACHE CAPACITY ALLOCATION METHOD BASED ON CLUSTERING

  • 摘要: 为合理分配信息中心网络(ICN)中每个节点的缓存空间,提出一种基于聚类的ICN缓存空间规划方法。根据节点属性数据集对ICN节点聚类后,以得到的聚类中心属性数据集为每一类分配缓存空间,后根据欧氏距离评估聚类效果为类内每个节点分配缓存空间,来提高缓存系统性能。在实际应用中,也可以为互联网服务提供商(ISP)提供一定的指导意义。仿真结果表明,该方法与均分分配部署方法、基于度数部署方法、基于介数部署方法相比,缓存命中率有明显提高,响应时延明显降低。

     

    Abstract: In order to reasonably allocate the cache space of each node in information-centric network (ICN), an ICN cache capacity allocation method based on clustering is proposed. After clustering the ICN nodes according to the node attribute data set, the obtained cluster center attribute data set was used to allocate cache space for each category. According to the Euclidean distance to evaluate the clustering effect, the cache space was allocated to each node in the category to improve the cache System performance. In practical applications, it can also provide certain guidance for internet service providers (ISP). The simulation results show that the proposed method improves the cache hit rate and significantly reduces the response latency compared with the equally distributed deployment method, the degree-based deployment method, and the betweenness-based deployment method.

     

/

返回文章
返回