Simulating computer systems: techniques and tools
Simulating computer systems: techniques and tools
Optimal load balancing in distributed computer systems
Optimal load balancing in distributed computer systems
Mobile IP: the Internet unplugged
Mobile IP: the Internet unplugged
Simulation Modeling and Analysis
Simulation Modeling and Analysis
Mobile IP; Design Principles and Practices
Mobile IP; Design Principles and Practices
Route Optimization for Mobile IP
Cluster Computing
Design and analysis of a replicated server architecture for supporting IP-Host mobility
ACM SIGMOBILE Mobile Computing and Communications Review
Hi-index | 0.00 |
In a continuously expanding mobile and wireless networking environment, mobile IP is the preferred standard responsible for providing Internet connectivity to roaming mobile nodes. Under certain conditions of traffic, such as in mobile IP networks supporting multimedia applications, overhead can cause delays at the mobility agents, i.e. foreign and home agents. While the mobile IP standard does not exclude the possibility of using multiple home agents, it does not impose any particular model either; therefore using only one home agent in a classic home network configuration can significantly bottleneck the IP data packet streams destined to the mobile nodes. In this paper, we develop a new simulator in order to evaluate the performance of a load balanced multiple home agents protocol extension, as well as several results obtained during a comprehensive simulation study of the system. We study several dynamic load balancing policies that distribute the traffic among several nodes in the home network, comparing the results obtained under different traffic shapes. We introduce a more realistic double-threshold load balancing policy, and compare its behaviour with that of other dynamic/static policies. Using simulation, we also analyze the impact of modifying the number of load balanced home agents on the overall system performance. The results show that some load balancing policies expected to perform better than others occasionally have an opposite behavior. This proves out to be highly dependent on the traffic pattern, especially its degree of burstiness.