IP Network Traffic Engineering

Albert Greenberg, AT&T Labs - Research
Managing large IP networks requires an understanding of the current traffic flows, routing policies, and network configuration. Yet, the state-of-the-art for managing IP networks involves semi-manual configuration of IP routers, and traffic engineering based on limited measurements. The networking industry lacks software systems that a large ISP can use to support traffic measurement and network modeling, the underpinnings of effective traffic engineering.

In this talk, we describe a unified set of software tools for managing the performance of IP backbone networks. The key idea is to generate global views of the network, on the basis of configuration and usage data associated with the individual network elements. Having created an appropriate global view, one can infer and visualize the network-wide implications of local changes in traffic, configuration, and control. A network provider can then experiment with changes in network configuration in a simulated environment, rather than the operational network. In addition, one obtains a sound framework for additional modules for network optimization and performance debugging. This talk is based on joint work with several others at AT&T Labs.

About the Presenter
Albert Greenberg works in the area of IP network management and performance, particularly on R&D for performance debugging and traffic engineering tools to help configure and manage operational IP networks.

HTML presentation
PPT presentation