Resource Provisioning for QoS support over IP-Networks

My PhD thesis can now be downloaded from here

A Scalable Framework for Traffic Policing and Admission Control at the Edge

Furies provides a control framework for scalable, efficient admission control and traffic policing. Furies leverages the knowledge of traffic demand distributions between ingress-egress pairs and the network topology within an ISP in making admission control decisions. We propose to aggregate admitted flows for policing at edge routers instead of monitoring individual flows. Furies achieves this by assigning a unique flow-identifier to every admitted flow based on its ingress and egress point. As a result, the amount of states maintained by the edge routers can be reduced from O(n) to O(square root(n)), where n is the number of admitted flows, while core routers are stateless. Simulation results show that we can successfully detect a majority (64-83%) of the malicious flows with virtually zero false-alarms without having to keep per-flow state at the edge. Our implementation demonstrates that Furies adds minimal processing overhead to edge routers and can be incrementally deployed.


Resource Provisioning via a Clearing House Architecture

The lack of a well-studied policy architecture to regulate resource provisioning within large domains or across multiple domains in a scalable manner has motivated our design of a Clearing House (CH) Architecture as an alternative solution. The CH attempts to provide better QoS assurance and higher network utilization, as offered by stateful networks (e.g., Int-Serv), while maintaining the scalability of a stateless network architecture (e.g., Diff-Serv).

Our CH architecture uses a hybrid of a flat and a hierarchical structure. A hierarchical structure helps in distributing the network state information among the various CH-nodes and reduces the amount of states maintained, while a flat structure is helpful for peer-to-peer provisioning across domains. At the top level, our architecture appears flat while the hierarchical structure is associated with large ISPs or ASs. We have also developed a distributed controller that attempts to maximize the effective throughput seen by the entire system and adapts to fluctuating load patterns. The CH-nodes close to the host networks are responsible for performing admission control. The edge routers maintain only aggregated state information about the flows and the core routers are completely stateless. The CH-nodes keep track of the intra- and inter-domain traffic patterns, and adapt aggregate reservations dynamically based on "Gaussian traffic predictors".

Details about the Clearing House design and simulation results are described in the following papers.


Network Resource Management for Latency-Sensitive Applications

The main objective of this work is to study the fundamental challenges and design criteria for future IP-based broadband network infrasturcture. Some IP-based real-time applications require some QoS guarantees, and therefore network designers have to go beyond "best effort". Our goal is to find a low-complexity extension to current network intelligence to manage the network resources more efficiently, and make the networks run better.
( Research Summary)


Misc. Projects

Infrastructure for a Secure Interface Between Wireless and Data Networks

CS261- Computer Securities Class Project, Fall 1998. (ppt, Project Report)
(Work with Mark D. Spiller)

Abstract
We have designed a security infrastructure and built a simple prototype that allows a wireless device such as a cel phone to interoperate securely with data networks. This would enable a valid user to use the handset to invoke multi-services such as unlocking an office door, turning on the lights, sending out emergency messages to campus police etc.
The required infrastructure would include support for authenticated, private messages/commands (for instance, voice-recognized key words or codes keyed into the keypad). Our study is based on GSM networks, and we address the following security issues:


Performance Evaluation of RSVP-enabled QoS Control

Summer internship project at Lucent Technology-UMTS, Swindon, UK, 1998.

Abstract
The main aim of the project is to demonstrate the feasibility and efficiency of providing IP services in UMTS over and IP-based backbone. As part of the effort, it is important to study the QoS control mechanisms that can support differentiated services in UMTS, especially for latency-sensitive applications such as real-time audio and video conferencing.
Resource reservation is essential for QoS provisioning in the Integrated Services Packet Networks (ISPNs) that we envision in the future UMTS networks. We set up a test-bed to evaluate effectiveness of RSVP-enabled QoS control for real-time traffic over a private subnet. Our preliminary performance study used both real-time audio/video streams, and simulated background traffic.


Relevant Links


Email chuah at ece.ucdavis.edu

[Home Page]