Research On The Consistency Of Mapping Cache In Identifier-based Universal Network

With the rapid development of information interconnection technology,Internet expands steadily,which brings more convenience to people,but also brings a variety of drawbacks due to the defects of the original design,such as the poor scalability,the poor security,poor mobility and other problems.In order to solve these problems,many organizations at home and abroad are committed to the research of next generation Internet to solve these problems.One of the main ideas is to split the identity and location information.National Engineering Laboratory for Next Generation Internet puts forward the concept of Universal Identifier Networks.It uses Access ID(AID)to indicate the identity of the terminal information,and uses Router ID(RID)to indicate the location of the terminal information.The Mapping System realizes the information storage and query functions.This article is based on one of Identifier-locator Separation Networks--Universal Identifier Networks.To address the problem of cache consistency caused by a mapping change,the existing distributed mapping system is improved.After improving the synchronization of mapping information's release and recovery process with interprocess communication,a mechanism to maintain strong cache consistency is proposed,which uses the server to realize the control of cache in access router.First this article introduces the background of Identifier-locator Separation Network's mapping system.Second this article introduces the principle of Distributed Mapping Service System of Universal Identifier Networks,the working entity and the analysis of the communication process.After improving the synchronization of mapping information's release and recovery process with interprocess communication,this paper proposed the design and the analysis of consistency module.On this basis,each consistency module is designed and implemented on the code.Functional tests are carried out on each function module,which verifies the automatic summarization and distribution of the domain information,and the timeliness of the cache.And we contrast the performance of a weak cache consistency scheme and the strong cache consistency scheme,propose the necessity of strong consistency.At last,the sixth chapter gives a summary of the whole work and proposes suggestions for further study work.