Worst-case delay bounds with fixed priorities using network calculus

Abstract

Worst-case delay bounds are an important issue for applications with hard-time constraints. Network calculus is a useful theory to study worst-case performance bounds in networks. In this article, we focus on networks with a fixed priority service policy and provide methods to analyze systems where the traffic and the services are constrained by some minimum and/or maximum functions: arrival/service curves. Our approach uses linear programming to express constraints of network calculus. Our first approach refines an existing method by taking into account fixed priorities. Then we improve that bound by mixing this method with other ones and provide a lower bound of the worst-case delay. Finally, numerical experiments are used to compare those bounds. Copyright © 2011 ICST.