A user level model for artificial internet traffic generation
The unprecedented growth of the internet and the rapid advances in computer networking technologies have raised important issues in the protocol design methodology. Today's protocol designers have an increasing need to test their protocols under realistic traffic load. While software simulation is a necessary step in the design pro-cess and can provide valuable information about the expected system characteristics, the existence of a tool that can go beyond the traditional simulation and is able to reflect closely real network traffic have become a great necessity. This need has led us to develop a new statistical model that takes into account the network traffic from the user level down to the application level, this thesis suggests a user level statistical model that utilizes traffic patterns that can closely mirror the expected user load without the necessity to have a real end user network for the test. Our model is a conversation level model and it uses a set of statistical and sofware tools to emulate the same type of traffic over the test network.
School Location:USA - Ohio
Source Type:Master's Thesis
Keywords:user level model artificial internet traffic generation load emulation
Date of Publication:01/01/2000