UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Traffic engineering methods for a network processor based multimedia router Elramly, Amr

Abstract

The thesis proposes methods to build network processor based multimedia router that provides more services than what is offered by the most expensive network routers. The new methods provide scalability to serve hundreds of thousands of multimedia streams and serve fast enough to meet the low latency required by the multimedia application. The first method proposed is a proxy agent that turns the router into multimedia aware router by preparing the profiling and the routing information for the forwarder and the scheduler, which simplifies the processes required for the rest of the router operation. The second method is Lowest Credit Weighted Fair Queuing or LCWFQ scheduler, that achieves a fairness index of 1 (One) and a work order of O(1) which is similar to the famous Deficit Round Robin (DRR) scheduling algorithm, except that the LCWFQ avoids the DRR weakness in implementation. The third method is an Adaptive Weight Adjustment (AWA) traffic shaper that the Internet Service Provider (ISP) may use when the total ISP commitments to the subscribers are more than what the network can handle. The AWA also serves the bursty streams when conflicting with the assigned SLA while maintains fair services to non-bursty traffic. The forth method is a three level memory index that fast search a record in the forwarding table where the number of memory accesses is always three no matter how large is the forwarding table. The thesis also includes two methods to translate the Service Level Agreement (SLA) into scheduling profiles and to report any bandwidth over-usage to help redefining the subscriber SLA. The proposed methods are designed to fit the network processors abilities to implement loosely coupled profiling, traffic shaping and scheduling.

Item Media

Item Citations and Data

License

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

Usage Statistics