新型互联网服务标识解析映射系统缓存机制研究

发布时间：2018-05-04 15:49

本文选题：新型互联网 + 服务标识　；参考：《北京交通大学》2017年硕士论文

【摘要】：随着互联网的不断发展,当今社会的各行各业对互联网的需求也是不断增加。为了减小用户的海量请求对网络设备造成的巨大压力,往往需要将用户的请求停止在服务器之前,即需要将服务器提供的部分服务做一份拷贝放置在离用户更近的位置来代替服务器工作,这就是缓存提供的功能。因此,高效的缓存技术便成了当下的研究热点,任何一个互联网系统在投入实际使用之前都需要有一套完善的缓存机制。在现有的新型互联网服务标识解析映射系统中,已经设计并实现了服务标识到接入标识的映射。但面对未来使用中对系统高可靠、低时延的需求,本文在其基础上,为其设计了一套完整的缓存方案。该缓存存在于映射系统的边缘,用于暂存通信对端的服务标识与接入标识的映射关系,以减少用户的查询时延,提高用户体验。本文设计的缓存系统主要着眼于以下几点。首先优化内存分配与划分。在涉及内存资源利用的项目中,往往不能在需要使用时直接进行手动分配,由于这样操作的内存不连续而导致大量的内存块无法得到利用,有限资源的利用率随即下降。在本文中需要设计一种高效的内存管理方案来减少内存冗余,使有限的资源得到最大化的利用;接着添加缓存替换算法。设备的内存往往都是有限的,设计缓存系统需要为其选择合适的替换算法来应对内存资源有限的问题;之后添加缓存一致性策略。为了满足新型互联网服务标识对缓存一致性的要求,选择合适的一致性策略来保证用户访问的数据的有效性;然后设计数据包格式和交互流程。为了使缓存系统完美地融入原有新型互联网服务标识解析映射系统中,需要设计所有缓存数据包的格式以及与原有网络实体之间的交互流程;最后编码实现所有的设计方案并应用于新型互联网服务标识解析映射系统。
[Abstract]:With the continuous development of the Internet, the demand for the Internet in various industries is also increasing. In order to reduce the huge pressure on network devices caused by users' massive requests, it is often necessary to stop the user's requests before the server. That is to say, it is necessary to make a copy of some of the services provided by the server to replace the server in a place closer to the user, which is the function provided by the cache. Therefore, efficient caching technology has become the current research hotspot. Any Internet system needs a set of perfect caching mechanism before it is put into practical use. The mapping from service identification to access identification has been designed and implemented in the existing new mapping system of Internet service identification. However, facing the need of high reliability and low delay in the future, this paper designs a complete cache scheme for the system. The cache exists on the edge of the mapping system and is used to temporarily store the mapping relationship between the service identity and the access identity of the communication peer, so as to reduce the query delay of the user and improve the user experience. The cache system designed in this paper mainly focuses on the following points. First, optimize memory allocation and partition. In the projects involving the utilization of memory resources, it is often impossible to allocate them manually when they need to be used. As a result of the discontinuity of memory in this way, a large number of memory blocks cannot be utilized, and the utilization rate of limited resources will then decline. In this paper, we need to design an efficient memory management scheme to reduce memory redundancy and maximize the utilization of limited resources, and then add cache replacement algorithm. The memory of the device is often limited, so it is necessary to select the appropriate replacement algorithm to deal with the problem of limited memory resources, and then add the cache consistency policy. In order to meet the requirement of cache consistency in new Internet service identification, we choose a suitable consistency policy to ensure the validity of the data accessed by users, and then design the data packet format and interaction flow. In order to integrate the cache system into the original new Internet service identity parsing and mapping system, it is necessary to design the format of all cached packets and the interaction process with the original network entities. Finally, all the design schemes are implemented by coding and applied to the new Internet service identity analysis and mapping system.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP393.09

【相似文献】