Response - Time Analysis for ATM Networks

The ability to guarantee timely delivery of messages is crucial in the design and validation of distributed hard real-time systems. The use of such systems is increasing, due to both the increased performance and reliability they can o er, and the need to control physically distributed equipment. As a consequence, methods to provide timeliness guarantees become more important. To support the guarantee of timely delivery, this thesis presents (1) a method to determine the worst case response-time for messages sent over Asynchronous Transfer Mode (ATM) networks, and (2) a method to determine how much bu er memory is needed in each ATM switch in the network. In addition, we present an associated priority assignment algorithm, which is essential for achieving e cient sharing of network resources. The presented methods are intended for call admission control of hard real-time tra c, i.e. they can be used to determine whether new tra c can be admitted or not. Our analysis methods are based on traditional response-time analysis for priority driven CPU scheduling, and they allow us to calculate end-toend response-times in ATM networks. In a comprehensive evaluation, we compare our method with other methods that provide similar guarantees. The evaluation shows that our response-time analysis enables higher utilisation of network resources compared to the other methods. We show that this success is a result of the high precision in our analysis, combined with the good performance of the priority assignment algorithm.