This dissertation proposes and evaluates a new approach for generating realistic traffic networking experiments. The main problem solved by our approach is generating closed-loop traffic consistent with the behavior of the entire set of applications in modern traffic mixes. Unlike earlier approaches, which described individual applications in terms of the specific semantics of each application, we describe the source behavior driving each connection in a generic manner using the a-b-t model. This model provides an intuitive but detailed way of describing source behavior in terms of connection vectors that capture the sizes and ordering of application data units, the quiet times between them, and whether data exchange is sequential or concurrent. This is consistent with the view of traffic from TCP, which does not concern itself with application semantics.
The a-b-t model also satisfies a crucial property: given a packet header trace collected from an arbitrary Internet link, we can algorithmically infer the source-level behavior driving each connection, and cast it into the notation of the model. The result of packet header processing is a collection of a-b-t connection vectors, which can then be replayed in software simulators and testbed experiments to drive network stacks. Such a replay generates synthetic traffic that fully preserves the feedback loop between the TCP endpoints and the state of the network, which is essential in experiments where network congestion can occur. By construction, this type of traffic generation is fully reproducible, providing a solid foundation for comparative empirical studies.
Our experimental work demonstrates the high quality of the generated traffic, by directly comparing traces from real Internet links and their source-level trace replays for a rich set of metrics. Such comparison requires the careful measurement of network parameters for each connection, and their reproduction together with the corresponding source behavior. Our final contribution consists of two resampling methods for introducing controlled variability in network experiments and for generating closed-loop traffic that accurately matches a target offered load.
Click here to get a PDF copy of this dissertation (5.8 MB uncompressed).
Last revised Mon Jun 9 9:37:44 EDT 2008 by jeffay at cs.unc.edu.