The scalability of the Internet hinges on our ability to tame the unpredictability associated with its open architecture. This project investigates the development of basic control strategies for reducing traffic burstiness and improving network utilization. Such strategies can be applied through Internet Traffic Managers (ITMs)---special network elements strategically placed in the Internet (e.g., in front of clients/servers or at exchange/peering points between administrative domains). We believe that the incorporation of such control functionalities will be key to the ability of the network infrastructure to sustain its own growth and to nurture the Quality-of-Service (QoS) needs of emerging applications.
In this project, we address the design of dynamic QoS control in programmable ITMs. The following are examples of basic capabilities that could be employed at different levels of the control architecture.
Differentiated Control enables ITMs to route flow aggregates with divergent characteristics on separate communication paths. Unlike traditional routing, we are examining routing metrics that respect both link-level and transport-level measures;
Aggregate Control enables ITMs to use congestion control mechanisms for collections of flows that share the same bottleneck. Unlike traditional congestion control, the objective is to manage "congestion-equivalent" flows as a set;
Proxy Control enables ITMs to filter out variability (e.g. loss) at shorter time-scales. Such a functionality is crucial for improving the stability and effectiveness of control mechanisms that operate over longer time-scales (e.g., end-to-end). Unlike traditional ad-hoc proxy approaches, the objective is to take into account the characteristics of the nested control loops that get formed between the ITMs and the end-systems.
The design is based on mathematical foundations from optimal control theory and statistical estimation. Specifically, functionalities at different levels of an ITM architecture are to be developed based on integrated control-theoretic models. These models would account for "nested" control loops that are driven by system characteristics, which are identified using statistical analysis of passive measurements. ITMs that are designed in such an integrated fashion, could increase flow throughput, reduce flow jitter and response time, and improve the stability, utilization, and scalability of the network.
We are implementing our QoS controls in a testbed deployed in a controlled local setting as well as over the Internet. Our testbed provides a programming interface to soft services, in which capabilities can be turned on or off and control parameters can be dynamically adjusted. The released software of the traffic controller will include tasks such as differentiated (traffic-aware) management, aggregate transmission control, wireless proxy for reliable transport, and measurement and characterization of control parameters.
itmBench, is now available.
This project is in part funded by the National Science Foundation under grant ANI-0095988 from the Special Projects in Networking program.