> By Guillaume Valadon and Clémence Magnien
With the Mobile IPv6 protocol, packets exchanged between two mobile nodes A and B must go through the home agent HA. The resulting communication distance between A and B is therefore the sum of the distances between A and HA, and between HA and B. Our goal is to find locations for home agents such that the communication distances are as short as possible. Regarding Mobile IPv6, the path length between two nodes A and B is not altered when the home agent is located on the shortest path between A and B. Consequently, good home agent locations correspond to vertices that belong to a large number of shortest paths in the graph.
Two different strategies are used to identify relevant locations for home agents: one uses the degree, and the other one the betweenness centrality. Giving a graph that represents a real communication network, we study home agents locations by placing them in vertices sorted by decreasing degree, and betweenness centrality, and comparing the distance between two vertices to the communication distances when Mobile IPv6 is used. These placement strategies are compared to a home agent located in the subnetwork that delivers the best performances.
When the home agent is located in one of the two vertices with the highest degree, or betweenness, the number of paths that are not modified is drastically increased. In the studied network, 30% and 34% of all shortest paths are not modified when the degree and the betweenness centrality are respectively used to select the home agent location. This is 3% with the best subnetwork.
In practice, there is a relation between high degree and high betweenness centrality. Indeed, it could be used by system administrators that want to deploy Mobile IPv6, but can not apply our graph based strategies because they do not have a graph model of their network. They could instead locate the home agent close to highly connected Network Operating Centers or routers; i.e. nodes with a high degree.
More information available in Guillaume Valadon’s PhD thesis.