This paper analyses the stability and fairness of two classes of rate control algorithm for communication networks. The algorithms provide natural generalisations to large-scale networks of simple additive increase/multiplicative decrease schemes, and are shown to be stable about a system optimum characterised by a proportional fairness criterion. Stability is established by showing that, with an appropriate formulation of the overall optimisation problem, the network's implicit objective function provides a Lyapunov function for the dynamical system defined by the rate control algorithm. The network's optimisation problem may be cast in primal or dual form: this leads naturally to two classes of algorithm, which may be interpreted in terms of either congestion indication feedback signals or explicit rates based on shadow prices. Both classes of algorithm may be generalised to include routing control, and provide natural implementations of proportionally fair pricing.

https://link.springer.com/content/pdf/10.1057%2Fpalgrave.jors.2600523.pdf

Rate control for communication networks: shadow prices, proportional fairness and stability

This paper addresses the issues of charging, rate control and routing for a communication network carrying elastic traffic, such as an ATM network offering an available bit rate service. A model is described from which max-min fairness of rates emerges as a limiting special case; more generally, the charges users are prepared to pay influence their allocated rates. In the preferred version of the model, a user chooses the charge per unit time that the user will pay; thereafter the user's rate is determined by the network according to a proportional fairness criterion applied to the rate per unit charge. A system optimum is achieved when users' choices of charges and the network's choice of allocated rates are in equilibrium.

/pdf/charging-and-rate-control-for-elastic-traffic-cumxyngz1i.pdf

Charging and rate control for elastic traffic

We propose an optimization approach to flow control where the objective is to maximize the aggregate source utility over their transmission rates. We view network links and sources as processors of a distributed computation system to solve the dual problem using a gradient projection algorithm. In this system, sources select transmission rates that maximize their own benefits, utility minus bandwidth cost, and network links adjust bandwidth prices to coordinate the sources' decisions. We allow feedback delays to be different, substantial, and time varying, and links and sources to update at different times and with different frequencies. We provide asynchronous distributed algorithms and prove their convergence in a static environment. We present measurements obtained from a preliminary prototype to illustrate the convergence of the algorithm in a slowly time-varying environment. We discuss its fairness property.

/pdf/optimization-flow-control-i-basic-algorithm-and-convergence-49tl2zuk8p.pdf

Optimization flow control—I: basic algorithm and convergence

In network design, the gap between theory and practice is woefully broad. This book narrows it, comprehensively and critically examining current network design models and methods. You will learn where mathematical modeling and algorithmic optimization have been under-utilized. At the opposite extreme, you will learn where they tend to fail to contribute to the twin goals of network efficiency and cost-savings. Most of all, you will learn precisely how to tailor theoretical models to make them as useful as possible in practice. Throughout, the authors focus on the traffic demands encountered in the real world of network design. Their generic approach, however, allows problem formulations and solutions to be applied across the board to virtually any type of backbone communication or computer network. For beginners, this book is an excellent introduction. For seasoned professionals, it provides immediate solutions and a strong foundation for further advances in the use of mathematical modeling for network design. (Less)

Routing, Flow, And Capacity Design In Communication And Computer Networks

This article reviews the current transmission control protocol (TCP) congestion control protocols and overviews recent advances that have brought analytical tools to this problem. We describe an optimization-based framework that provides an interpretation of various flow control mechanisms, in particular, the utility being optimized by the protocol's equilibrium structure. We also look at the dynamics of TCP and employ linear models to exhibit stability limitations in the predominant TCP versions, despite certain built-in compensations for delay. Finally, we present a new protocol that overcomes these limitations and provides stability in a way that is scalable to arbitrary networks, link capacities, and delays.

/pdf/internet-congestion-control-cxss2y6xaf.pdf

Internet congestion control

A quality of service guarantee for voice and other delay sensitive transmissions within an Internet Protocol (IP) network is provided by identifying the IP network path utilized for IP packet transmission between source and destination edge devices and virtually provisioning IP network path bandwidth for priority voice traffic. Priority for voice packets and admission control of new voice calls (and other delay sensitive traffic) based on the remaining available capacity over the IP network path guarantees that high priority voice (and other delay sensitive traffic) meet stringent delay requirements. A Virtual Provisioning Server is utilized to maintain bandwidth capacity data for each path segment within the IP network and to forward the bandwidth capacity data to a Signaling Gateway. The Signaling Gateway determines whether to accept or reject an additional delay sensitive traffic component based upon available bandwidth capacity for an IP network path. The Signaling Gateway then signals the originating source edge device as to its determination to accept or reject. Quality of Service guarantees concerning acceptable delay and jitter characteristics for real-time transmission over an IP network are therefore provided without the need to directly signal the individual IP routers over which an IP network path is established.

Method for providing quality of service for delay sensitive traffic over ip networks

Software defined networking and network function virtualization are widely deemed two critical pillars of the future service provider network. The expectation for significant operations cost savings from network programmability, open APIs, and operations automation is frequently mentioned as one of the primary benefits of the NFV/SDN vision. Intuitively, the flexibility and simplification values attributed to NFV/SDN lead the industry to the conclusion that operating expenses will decrease. This article provides a view into the operational costs of a typical service provider and discusses how the NFV/SDN attributes can be expected to influence the business equation. The drivers of OPEX change, the directionality of the change, and the categories of OPEX most affected based on our analysis from interactions with a number of service providers worldwide are presented in a structured analysis.

How will NFV/SDN transform service provider opex?

The current state of the art in digital broadband access technologies to support emerging telecommunications services makes imminent the introduction of interactive broadband services-including data, video and the Internet-into the residential market. Over the last few years, much attention has been paid to the development of media access control protocols for cable TV networks that will allow the immediate support of broadband data services as the first step toward enhanced communications services for residential users. We review some of the architectural options that must be carefully considered in order to deliver Internet protocol (IP) services to such users in an efficient yet flexible manner. Future residential cable data services are expected to deliver Internet access, work-at-home applications, small business access, local area network LAN-LAN interconnect, and LAN emulation services over cable (CATV) networks.

/pdf/architectures-for-broadband-residential-ip-services-over-4huqgfvi25.pdf

Architectures for broadband residential IP services over CATV networks

Over the last couple of years both the ATM Forum and ITU-T have converged on a closed-loop, feedback based rate control mechanism as the framework for the Available Bit Rate ATM Transfer Capability. In this framework, an ABR source adapts its transmission rate to changing characteristics of the ATM network by either testing periodically the state of the network or through direct feedback from the network. The state of the network is conveyed to the source either as a simple binary feedback signaling congestion or no congestion, or as an explicit rate indicating the rate at which the source can transmit. In this paper we review the detailed framework for the ABR service standardized by the ATM Forum, the underlying service philosophy and protocol design goals for ABR flow control, and the premises and high-level performance characteristics for the more promising ABR rate control schemes proposed so far.

Rate control algorithms for the ATM ABR service

Many popular point-to-point data link protocols use the high-level data link control (HDLC) framing mechanism, which delineates protocol data units (PDUs) by means of a special bit pattern or flag. When such a flag occurs in the payload portion of a frame, an escape byte used to pad the transmitted byte stream enables the receiver to differentiate between a true framing flag and an occurrence of the flag pattern in the user information. The need to process each byte in the incoming byte stream to identify the flag pattern makes this frame delineation method increasingly more complex and expensive to implement as the interface speed increases. In addition, the byte stuffing operation performed when the flag pattern appears in the user information stream expands the carried traffic, interfering with quality of service (QoS) management procedures and making the link susceptible to malicious attacks. These factors limit the scalability and QoS management capabilities of the flag-based data link protocols, particularly at high link rates. The increasing popularity of the packet over SONET (POS) protocol stack makes it desirable to find alternative framing protocols. In this paper, we present a simple data link (SDL) protocol. Based on using a Length Indicator field and a header cyclic redundancy check (CRC), rather than a flag, to delineate frames, SDL is inherently scalable to high speeds and provides constant transmission overhead. We discuss the error control, delineation, and resynchronization mechanisms used by SDL, and we describe a link scrambler designed to protect against malicious attacks that attempt to generate very low bit transition density on the line. We also present additional link control functions designed to make SDL suitable for next-generation multiservice networks.

Enrique Hernandez-Valencia

Papers

Method for providing quality of service for delay sensitive traffic over ip networks

How will NFV/SDN transform service provider opex?

Architectures for broadband residential IP services over CATV networks

Rate control algorithms for the ATM ABR service

A simple data link protocol for high-speed packet networks