Multiservice Packet Networks Research Articles

Capacity assignment in multiservice packet networks with soft maximum waiting time guarantees

We address the problem of capacity assignment in packet networks with soft maximum waiting time guarantees. We present a multi-level framework for designing and analyzing capacity assignment methods in multiservice networks. Moreover, we propose a novel capacity assignment method that provides a tradeoff between capacity requirements and quality for the voice traffic class. The proposed method aims at providing slightly softened quality guarantees with some controlled tolerance to service degradation in order to notably reduce the amount of required capacity. In order to demonstrate the key characteristics of the proposed method, derivations, analysis, and results are presented over multiple levels of detail starting at the buffer level and extending to the node level, the path level, and finally the network level. At the path and network levels, the problem of capacity assignment with soft maximum waiting time guarantees is formulated as an optimization problem in order to obtain minimized values of the link capacities. Numerical calculation results are presented based on example scenarios at the different levels in order to demonstrate the effectiveness of the proposed method compared to other approaches. Moreover, Monte Carlo simulation results are presented whenever necessary in order to validate assumptions and further support observations.

A meta-model for multiobjective routing in MPLS networks

MPLS (Multiprotocol Label Switching) enables the utilisation of explicit routes and other advanced routing mechanisms in multiservice packet networks, capable of dealing with multiple and heterogeneous QoS (Quality of Service) parameters. Firstly the paper presents a discussion of conceptual and methodological issues raised by multiobjective routing optimisation models for MPLS networks. The major contribution is the proposal of a multiobjective routing optimisation framework for MPLS networks. The major features of this modelling framework are: the formulation of a three-level hierarchical routing optimisation problem including network and service performance objectives, the inclusion of fairness objectives in the different levels of optimisation and a two-level stochastic representation of the traffic in the network (traffic flow and packet stream levels). A variant of the general model for two classes of traffic flows, QoS traffic and Best Effort traffic, is also presented. Finally a stochastic teletraffic modelling approach, underlying the optimisation model, is fully described.

Round-Robin with VirtualClock Scheduling Algorithm in Multiservice Packet Networks

In this paper, we present a scheduler that incorporates round robin service within a VirtualClock discipline. Time-stamp based scheduling algorithms attain a low local delay bound and performance guarantee, but are computationally complex. On the other hand, round robin schemes are simple to implement and have computational complexity of O(1) but they are well known for their output burstiness and short-term unfairness. In order to overcome this problem, we combine round robin with VirtualClock in an algorithm we call VCRR. VCRR possesses better fairness than simple round robin, low jitter and a good scheduling delay bound. At the same time, VCRR preserves the O(1) time complexity of round robin. Simulation experiments show VCRR's efficiency in terms of delay performance, jitter and fairness..

A framework for end-to-end deterministic-delay service provisioning in multiservice packet networks

The problem of providing end-to-end delay guarantees for deterministic-delay services in multiservice packet networks is addressed through a combination of dynamic resource reservation and routing. Our model is based on using rate-controlled earliest-deadline-first (RC-EDF) for providing hard bounds on end-to-end delays. With RC-EDF, a certain delay bound has to be allocated for a connection at each node in the selected path. The most commonly used resource reservation policy is uniform allocation which is based on dividing the end-to-end delay bound equally among the nodes in the selected path. This simple allocation policy could lead to nonuniform resource loading and subsequently lead to high blocking rates. Moreover, the most commonly used routing method is shortest-path first routing which is known to lead to network hotspots. We propose a set of dynamic nonuniform resource reservation policies and dynamic routing methods. One of the routing methods is the well-known widest-shortest path method and the other is a dynamic routing method that adaptively adjusts link costs and uses a similar algorithm to shortest-path routing (e.g., Dijkstra's algorithm). We show that for both uniform and nonuniform traffic loading of some example network topologies that the combination of the proposed resource reservation policies and dynamic routing can lead to significant reduction in the connection blocking ratio in all loading conditions except for excessively high loads.

Wireless service provider backbone network evolution with packet technologies

Moving toward the third-generation (3G), wireless service providers (WSPs) have key decisions to make in evolving their backbone networks in a cost-effective way. Today, most WSPs in the United States deploy fully meshed, circuit-switched metro backbone networks and hand off long-distance traffic to an interexchange carrier (IXC). With the increased complexity of managing a growing meshed network and the emergence of enhanced data services, the goal is to evolve toward a multi-service packet network. This paper presents a rigorous model for circuit-to-packet (C2P) network evolution planning that addresses evolution alternatives for voice services, technology choice, network architectures, design, performance, reliability, and financial analysis. The model allows WSPs to quantify the benefits of various evolution scenarios and helps Lucent Mobility Solutions assess product roadmaps. The paper also presents a case study based on typical code division multiple access (CDMA) WSPs in North America that examines various C2P evolution strategies in metro, regional, and national backbones based on different choices of technology (TDM vs. ATM vs. IP), product availability, network architectures, and traffic forecasts. Also included are key insights and a summary of the study results.

Performance Evaluation of Resource Reservation Policies for Rate-Controlled Earliest-Deadline-First Scheduling in Multi-Service Packet Networks

The paper addresses the issue of reserving resources at packet switches along the path of flows requiring a deterministic bound on end-to-end delay. The switches are assumed to schedule outgoing packets using the Rate-Controlled Earliest-Deadline-First (RC-EDF) scheduling discipline. EDF is known to be an optimal scheduling discipline for deterministic delay services in the single scheduler case. We propose a number of static and dynamic reservation policies for mapping the end-to-end delay requirement of a flow into local delay deadlines to be reserved at each scheduler. These policies are based on non-even resource reservation where the resources reserved depend on the capacities and loading at each node in the network. We define and prove the optimality of a certain non-even policy for the case of a single path network with homogenous static traffic. We present extensive simulation results for different scenarios which show that dynamic non-even resource reservation provides superior performance when compared to simple policies such as even dividing of end-to-end delay among the schedulers.

Bandwidth overhead of probe technology guaranteeing QoS in packet networks

IP and MPLS traffic engineering in multiservice packet networks depend on the accuracy of the performance monitoring capabilities that network operators and service providers also use to monitor and maintain their service level agreements. It is found that the feasibility of using active packet probing is highly dependent on the variability of the packet traffic characteristics, and the desired level of accuracy. Bursty traffic may require an unrealistically high number of probes, resulting in unacceptable bandwidth overheads even to provide an accurate estimate of mean packet delay.

The structure and management of service level agreements in networks

The paper proposes a structure for quality-of-service (QoS)-centered service level agreements (SLA), and a framework for their real-time management in multiservice packet networks. The SLA is structured to be fair to both parties, the service provider and their customer. The SLA considered here are for QoS assured delivery of aggregate bandwidth from ingress to egress nodes; however, the control and signaling is for the more granular flows or calls. A SLA monitoring scheme is presented in which revenue is generated by the admission of flows into the network, and penalty incurred when flows are lost in periods when the service provider is not SLA compliant. In the SLA management scheme proposed, the results of a prior off-line design are used, in conjunction with measurements taken locally at ingress nodes, to classify the loading status of routes. The routing and resource management are based on virtual partitioning and its supporting mechanism of bandwidth protection. The effectiveness of SLA management is measured by the robustness in performance in the presence of substantial diversity in actual traffic conditions. A simulation testbed called D'ARTAGNAN has been built from which we report numerical results for a case study. The results show that the SLA management scheme is robust, fair and efficient over a broad range of traffic conditions.

SRR: An O(1) time complexity packet scheduler for flows in multi-service packet networks

In this paper, we present a novel fair queueing scheme, which we call Smoothed Round Robin (SRR). Ordinary round robin schedulers are well known for their burstiness in the scheduling output. In order to overcome this problem, SRR codes the weights of the flows into binary vectors to form a Weight Matrix, then uses a Weight Spread Sequence (WSS), which is specially designed to distribute the output more evenly, to schedule packets by scanning the Weight Matrix. By using the WSS and the Weight Matrix, SRR can emulate the Generalized Processor Sharing (GPS) well. It possesses better short-term fairness and schedule delay properties in comparison with various round robin schedulers. At the same time, it preserves O (1) time complexity by avoiding the time-stamp maintenance employed in various Fair Queueing schedulers. Simulation and implementation experiments show that SRR can provide good average end-to-end delay for soft real-time services. SRR can also be implemented in high-speed networks to provide QoS for its simplicity and low time complexity.

A simple data link (SDL) protocol for next generation packet network

The simple data link (SDL) is a new framing protocol for variable/fixed length packets over a general-purpose point-to-point communications channel. SDL extends the HEC-like framing mechanism used in ATM to variable-length data. Its low implementation complexity makes it particularly suitable for high-speed transport links in wavelength channels and dark fiber applications. SDL is a very attractive alternative to conventional solutions such as ATM and PPP-over-SONET (POS) for high data rate environments. SDL has also been designed to facilitate many OAM&P functions needed in next generation multiservice optical packet networks including multiprotocol encapsulation, virtual links, quality-of-service (QoS) differentiation, link management, and control. We describe framing and data link synchronization procedures in SDL, and evaluate its performance over octet and bit synchronous transport facilities.

An expert system-based approach to capacity allocation in a multiservice application environment

The problem of bandwidth allocation for multiservice packet networks is discussed, and an expert system solution approach for solving it is described. The approach requires a minimal amount of processing, is very efficient, and has a fast response time. The implementation is based on modular capacity expansions and a priori limits on the number of connections for each service type. This approach can perform near-optimal capacity allocation in real time by using heuristic knowledge to virtually eliminate the required computations in this process. This implementation strategy ensures that each service demand meets its objectives. Concurrency control of this distributed expert system is considered. The advantages of the implementation in providing the ability to make rational decisions in an environment of uncertainty and imprecision, such as in the presence of nonsystematic overloads and network failures, are discussed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Error control techniques for integrated services packet networks

The design of a multiservice packet network must ensure that delays to speech packets are minimized while data and other types of packets are delivered without error. The author suggests some ways in which error detection, forward error correction (FEC), and automatic repeat request (ARQ) schemes may be utilized in integrated services packet networks (ISPNs) to ensure that satisfactory error performance and reliability standards are achieved. The results show that ARQ schemes combined with reliable error detection are the most practical way of achieving reliable error control in integrated services networks. Also, such error control schemes can have performance advantages if applied on a region-by-region basis rather than simple end-to-end.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>