Research And Optimization Of Local Routing Mechanisms In Network Simulation

Network simulation plays a vital role in attempting to characterize both the network performance and the network design. In research of the network simulation, routing is the most important among many factors that affect simulation scale and efficiency. How to simulate the routing behaviors of network faithfully and efficiently is a key problem in condition of limited hardware resources. Until now all the current routing mechanisms have their own disadvantages. After analyzing some of the current typical routing mechanisms, the dissertation presents an efficient routing compulation and storage approach, named TRMTree_Nix.TRMTree_Nix takes terminal router as study object to compute and store the routing information. And the threshold is combined with a calculation steps variant to stabilize and improve the coverage of the minimum spanning tree. Experimental results show that the TRMTree_Nix routing calculation and storage mechanism can reduce the storage space and the routing pre-computation time effectively. When making routing lookup, TRMTree_Nix use static routing table to store most of the shortest-path routing information, and for those that can't be stored by static routing table, we use Nix-Vector routing to compute them on demand. Performance comparison and worm behavior experiments show that the TRMTree_Nix routing lookup mechanism achieves higher performance, and is much more appropriate for the simulations of complex applications on large-scale topologies.When the effects of topology changes need to be considered in the Network simulation application, the dissertation proposes the LUMTree_Nix routing lookup mechanism on TRMTree_Nix which can sense changes in dynamic routing topology. Based on the static routing table and the queue of topology changes, with the comparation of current time and the awareness of the node, we can get the right routing state to meet the topology changes requirements.