Name And Address Separation Of The Hierarchical Design Of The Mapping System Under Study

With the continued expansion of the Internet, IP address's semantic overloading leads the problems in mobility, scalability, security, and makes the core network BGP routing table entries rapidly growth, and limits the traffic engineering, multi-homing and other new technologies. It is believed that identifier/locator split system may be used to solve these problems. The key of the system is the identifier-to-locator's mapping. In this paper, we firstly classify the mapping system according to their structures, and then compare their performances. Finally, some future research problems are discussed.This thesis is commenced around the mapping system of the identifier/locator split system. The mapping system must resolve the querying request from the nodes efficiently. High querying efficiency and in the same time, scalability, security and reliability of the mapping system must be guaranteed too. This thesis conducts a detailed description and simulation for two mapping systems LISP-CONS and LISP-DNS of the identifier/locator split system. In this paper, by building simulation models, we found that the mapping system with hierarchical structure has low design complexity and advantage in terms of the querying efficiency and scalability.The first chapter introduces the research background of the identifier/locator split system and the difficult points,and summarizes three separate solutions HIP, HAIR, and LISP.The design of the mapping system is the key point to the identifier/locator split system.The second chapter describes three designs LISP-CONS, LISP-DNS and LISP-DHT, including their structure and querying processes.In the third chapter from mobility, scalability, reliability, table size,and so on, the three designs LISP-CONS, LISP-DNS and LISP-DHT were respectively compared and analyzed.Chapters IV and V are the OPNET simulation of the LISP-CONS and LISP-DNS, in order to make results more realistic, the source data of the simulation is from the various Chinese provinces, cities,the actual IP number and the population. And at the same time, this thesis designs different simulation programs. By comparing the results of the experiments,this thesis points out the the shortcomings of two mapping systems based on the original, and accordingly proposes some improvements for the two mapping systems.The sixth chapter summarizes the full text, and points out the research direction in the future.