NeTS-NR: Stochastic Network Calculus: Theory and Tools for the Analysis of Future Networks and Applications

University Of Virginia Main Campus, Charlottesville VA

Investigators

Abstract

Currently, the development of revolutionary new network approaches is ham-pered by the lack of methods to evaluate the performance of radically different designs of network architectures, protocols, and applications. The goal of this project is to explore new theoretical concepts and algorithms for predicting the delay and throughput performance of future networks. The project develops a stochastic network calculus that leads to simple models and fast computational methods for networks that are very different from the networks and protocols used today. The research team will determine the capabilities and limitations of this approach and the project will address applications to scheduling in high-speed networks, to the verification of service level agreements, and in the analy-sis of feedback-based buffer management and congestion control algorithms. Efficient computational algorithms derived from the theory permit other re-searchers to use the methods of the calculus without requiring familiarity of the underlying theory. This project presents urgently needed methods for the analysis of the communication infrastructure of the 21st century and it stimu-lates research in analytical methods in science and engineering.