Computer Science Department News

There will be a Colloquium on Friday, March 18, 2011 at 10:30 in the E&CS Building first floor auditorium entitled "Towards Traffic Benchmarks for Empirical Networking Research: The Role of Structure in Modeling Application Data Exchanges" by Jau Aikat.

Dr. Jay Aikat is a postdoctoral researcher in the Department of Computer Science at the University of North Carolina at Chapel Hill (UNC-CH).

The following is the abstract for the Colloquium:

With the rise of phenomena such as social networks, cloud computing, and wireless devices, the number of different types of applications that constitute network traffic keeps constantly changing and ever increasing. This has compounded the problem of modeling these applications for realistic empirical evaluations.

Network testbeds and simulators remain the dominant platforms for evaluating networking technologies today. The results of such experiments are valid only to the extent that the traffic generated to drive these experiments accurately represents the traffic carried in real production networks.

In this talk, I shall discuss our research, demonstrating how certain components of network traffic, such as the structure of application data exchanges, can have a larger impact on the results obtained through experimentation than other dimensions of traffic such as round-trip time. Such findings point to the pressing need for traffic benchmarks in networking research; such benchmarks are common in many other areas of computing.

Through testbed experiments performed with synthetically generated network traffic, emulating a large number of applications from two very different traffic sources, and using several models of application data exchanges (ADEs), we demonstrate the strong effects of ADE modeling on performance measures such as queue length at routers, number of active connections in the network, user response times, and connection durations. We intend to use these results to advance the discussion of standards for empirical networking research, and to create a traffic benchmark suite, refreshing it as networked applications evolve.