Abstract
This chapter describes various aspects of deterministic network analysis. The events and processes in packet multiplexer models are presented. The chapter introduces some general notations that are used repeatedly in the context of traffic flows through packet networks. It develops a deterministic network calculus for packet networks. This calculus takes as inputs deterministic bounds on the incoming traffic and a characterization of network elements via service curves. As outputs, this calculus produces characterizations of departure processes and provides end-to-end delay bounds. A key input–output relationship is provided by the (min,+) convolution, much like the (+, x) convolution of linear systems theory. The given traffic processes can be made to conform to deterministic bounds (or envelopes) by passing them through regulators. The most popular traffic regulator is the leaky bucket, which Is analyzed in detail. Scheduling is the mechanism by which desired service curves are obtained for the various traffic flows sharing a link. Weighted fair queueing (WFQ) is discussed in detail and a performance bound that relates the packet delay in WFQ to that in generalized processor sharing (GPS) is proven. Finally, the chapter shows how some of the calculus developed in it is proposed to be used in the IntServ proposals for the provision of QoS in the Internet. The deterministic traffic models and network calculus are elaborated. The service curves for network elements are also analyzed.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.