Distributed simulation of large-scale and detailed models

In this paper, we present a new approach for the distributed simulation of large scale and detailed models. Our approach, based on the migration of simulated entities, enhances the simulator speed jointly addressing two main problems of distributed simulation: the reduction of the communication overhead and the load-balancing in the distributed execution architecture. The proposed method dynamically reconfigures the simulation, taking care of the performance of each part of the execution architecture and dealing with the unpredictable fluctuations of the available computation power and the communication load of each execution unit. In this way, low-cost commercial-off-the-shelf hardware can be used to run fast and cost-effective distributed simulations. A fine-grained model of the 802.11 DCF protocol has been used for the performance evaluation of the proposed approach. The results demonstrate that it is feasible for the detailed simulation of very large scale models such as wirele ss networks.