未来信息网络下智慧缓存策略研究与实现

发布时间：2018-06-01 23:00

本文选题：内容中心网络 + 分类　；参考：《北京交通大学》2014年硕士论文

【摘要】：摘要：随着人们越来越依赖互联网获取资源和信息,互联网所承受的压力与日俱增。从用户角度来讲,相同区域的用户在网络数据传输上存在着时间和地域的相关性,当区域中的用户访问某个数据后,该区域的其他用户也很有可能再次访问该数据,造成同一数据的重复传输,浪费网络带宽。网络缓存技术的出现在一定程度上缓解了以上难题。随着人们对网络的认识由基于主机位置(IP地址)到基于网络内容的逐渐转变,网络缓存的重要性逐渐凸显。在信息中心网络中,现有的缓存策略相对比较僵化,不能动态地适应不同用户的需求和信息中心网络下高缓存性能的要求。本文通过对信息中心网络下的现有缓存技术进行分析,提出了一种在CCN中基于分类映射的缓存管理策略。该缓存管理策略使路由器能对网络中的流量和区域用户进行分类和管理,更加符合不同区域的用户对网络资源的不同需求。该方案能够支持建立资源类别和用户类别的缓存映射关系表进行工作,并通过建立在缓存映射关系表上的缓存策略对缓存内容进行管理,提高资源利用率,优化用户体验。在此缓存管理策略基础上,本文提出了基于内容中心网络的智慧缓存系统,其主要包括内容分类单元,用户分类单元,缓存建立单元,缓存维护单元,缓存决定策略执行单元和缓存替换策略执行单元。其中,内容分类单元通过分析数据包对网络中的内容进行分类,用户分类单元通过分析数据包对网络中的用户进行分类。经过以上两功能单元得到的结果,通过缓存建立单元的分析建立起了用户类别与内容类别的缓存映射结构。缓存维护单元负责维护已经建立好的缓存映射结构。缓存决定策略执行单元和缓存替换策略执行单元共同作用于缓存映射结构,对网络内容进行限制和管理。
[Abstract]:Absrtact: as people rely more and more on the Internet to obtain resources and information, the pressure on the Internet is increasing day by day. From the user's point of view, users in the same area have a temporal and geographical correlation in the transmission of network data. When the user in the region accesses the data, it is very likely that the other users in the region will also access the data again. It causes repeated transmission of the same data and wastes network bandwidth. The emergence of network caching technology alleviates the above problems to a certain extent. With the gradual change of people's understanding of network from host location to network content, the importance of network cache is becoming more and more important. In the information center network, the existing cache strategy is relatively rigid and can not dynamically meet the needs of different users and high cache performance requirements under the information center network. Based on the analysis of the existing cache technology in the information center network, this paper proposes a cache management strategy based on classification mapping in CCN. The cache management strategy enables routers to classify and manage network traffic and area users, which is more consistent with the different requirements of users in different regions for network resources. This scheme can support the establishment of cache mapping relation tables of resource category and user category, and manage the cache content through the cache policy based on the cache mapping relation table, improve the resource utilization and optimize the user experience. On the basis of this cache management strategy, this paper proposes an intelligent cache system based on content center network, which includes content classification unit, user classification unit, cache establishment unit and cache maintenance unit. The cache determines the policy execution unit and the cache replacement policy execution unit. The content classification unit classifies the content in the network by analyzing the data packets, and the user classification unit classifies the users in the network by analyzing the data packets. After the results obtained by the above two functional units, the cache mapping structure of user category and content category is established by analyzing the cache establishment unit. The cache maintenance unit is responsible for maintaining the cache mapping structure that has been established. The cache decision policy execution unit and the cache replacement policy execution unit act together on the cache mapping structure to restrict and manage the network content.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.07

【参考文献】