Research On Network Modeling Based On Network Calculus

With the technological advances, communication systems have been greatly revolutionized. The advent of high speed networking has introduced opportunities for some demanding applications. These applications have stringent performance requirements in terms of throughput, delay, jitter, and loss rate. So, IETF proposed some guaranteed service schemes aiming at providing guaranteed services for various applications, and one of the most important is IntServ. Today's packet-switching integrated-services networks provide the communication infrastructure for the Internet. Unlike traditional communication networks which are designed to offer a single type of service, integrated-services networks are able to offer multiple services for various types of traffic, such as data, voice, video and so on. So integrated-services networks have to support real-time communication services that allow clients to transport information with performance guarantees expressed in terms of delay, jitter, throughput, and loss rate. However, the analysis of performances, especially the end-to-end performances has been an open complex problem.Network Calculus is a set of recent developments that provide deep insights into flow problems encountered in networking. The foundation of network calculus lies in the mathematical theory of Min-Plus algebra. With network calculus, we are able to understand some fundamental properties of integrated-services networks, window flow control, scheduling and buffer or delay dimensioning. By applying network calculus, the maximum end-to-end delay bound, the maximum backlog bound, jitter and effective bandwidth of IntServ guaranteed services can be derived.In this paper we research deeply on the method of modeling and analysis networks based on network calculus. We first introduce the basic network calculus concepts. Two of the most important analysis tools are arrival curves and service curves. Then the concepts of leaky-bucket, generic cell rate algorithm (GCRA), constant bit rate (CBR) flow, variable bit rate (VBR) flow are described in the appropriate framework, of arrival curves. We show that all rate-based packet schedulers can be abstract by a simple rate latency service curve. And by applying these fundamental rules of network calculus, bounds on delay, buffer and effective bandwidth for leaky bucket, GCRA, CBR and VBR can be derived and some practical examples are given. We also discuss application of network calculus to resource reservation, admission control and performance analysis in IntServ networks.To address the performance issues of service disciplines, In this paper, we also apply the new analysis technique of network calculus approach. We survey several service disciplines that are proposed in the literature to provide per-connection end-to-end performance guarantees in packet-switching networks. We use one of the basic network calculus tools, known as service curve, to model service disciplines. We derive their performance guarantees they can provide with the basic analysis tools of network calculus; We also describe their mechanisms, their similarities and differences. Various issues and tradeoffs in analyzing service disciplines for guaranteed performance service are discussed, and a general framework for studying and comparing these disciplines are presented.The validity and effectiveness of the analytic results are verified with application examples. The results presented here promise to be applicable for network design and optimization.