When designing a network simulation environment intended specifically for modeling large-scale topologies, a number of issues must be addressed by the simulator designer. Memory requirements for network simulation engines can grow quadratically with the size of the simulated topology and can easily exceed available memory on modern workstations. The number of outstanding simulation events grows linearly with the number of packets in flight being modeled, and can lead to performance bottlenecks when managing a sorted event list of millions of events. Tracking the results of the simulation using a packet-level log file can result in excessive usage of disk space. We discuss the design of the Georgia Tech Network Simulator (GTNetS) with emphasis on how GTNetS addresses these issues. We give results from performance experiments showing the reduction in memory and event list size as a result of our design decisions.

Simulations are used extensively for studying artificial intelligence. However, the simulation technology in use by and designed for the artificial intelligence community often fails to take advantage of much of the work by the larger simulation community to produce distributed, repeatable, and efficient simulations. We present the new system known as System for Parallel Agent Discrete Event Simulator, (SPADES), which is a simulation environment for the artificial intelligence community. SPADES focuses on the agent as a fundamental simulation component. The thinking rime of an agent is tracked and reflected in the results of the agents' actions by using a Sofnvare-in-theb o p mechanism. SPADES supports distributed execution of the agents across multiple systems, while at the same time producing repeatable results regardless of network or system load. We discuss the design of SPADES in detail and give experimental results. SPADES is flexible enough for a variety of application domains in the artificial intelligence research community.

In this paper we evaluate the Enhanced Interior Gateway Routing Protocol (EIGRP) via packet simulations. EIGRP, an intra-domain routing protocol developed by Cisco, is mainly based on the Diffusing Update Algorithm (DUAL) which computes shortest paths distributedly without creating routing-table loops or incurring counting-to-infinity problem. Previous studies showed EIGRP's ability to adapt quickly to routing changes in medium-scale networks. In our research, we developed a detailed simulation model of EIGRP (publicly available), and we used it to evaluate EIGRP performance under a very dynamic network. Our results showed that EIGRP converges faster than a single TCP timeout in most cases. The simulated network was a composite of wired and wireless hosts, and the results hold for both types of media. In addition, the study showed a feasible approach for seamless mobility and continuous connectivity for users of mobile wireless devices as they move within an Autonomous System (AS).

No abstract