GloMoSim: a library for parallel simulation of large-scale wireless networks
PADS '98 Proceedings of the twelfth workshop on Parallel and distributed simulation
On the accuracy of MANET simulators
Proceedings of the second ACM international workshop on Principles of mobile computing
Comparison of TCP Reno and Vegas in wireless mobile ad hoc networks
LCN '00 Proceedings of the 25th Annual IEEE Conference on Local Computer Networks
On credibility of simulation studies of telecommunication networks
IEEE Communications Magazine
| Hi-index | 0.00 |
Network researchers and designers rely on network simulation tools to do performance evaluation. There are a number of different network simulation tools, each with a set of publicly available models. Two of the more popular tools, ns-2 and Opnet have a large library of available models. The existing models can be used to configure simulation experiments in a relatively quick and easy manner. The choice of simulation tool often depends on the availability of models as well as the user's preference. Most users assume that performance results on different simulators will be comparable, particularly when using models that have been widely used and tested. Using simple simulation experiments, this paper shows that TCP performance results obtained from ns-2 and Opnet can have significant differences. The differences are exposed when TCP is tested with error rates ranging from 1% packet loss to 20% packet loss. A study of the actions taken by TCP as a result of the losses provides insight into the performance differences.