This paper presents a control-theoretic approach to reactive flow control in networks that do not reserve bandwidth. We assume a round-robin-like queue service discipline in the output queues of the network’s switches, and propose deterministic and stochastic models for a single conversation in a network of such switches. These models motivate the Packet-Pair rate probing technique, and a provably stable rate-based flow control scheme. A Kalman state estimator is derived from discrete-time state space analysis, but there are difficulties in using the estimator in practice. These difficulties are overcome by a novel estimation scheme based on fuzzy logic. We then present a technique to extract and use additional information horn the system to develop a continuous-time system model. This is used to design a wuisnt of the control law that is also provably stable, and, in addition, takes control action as rapidly as possible. Finally, practical issues such as correcting parameter drift and cmmlination with window flow control are described.

In the Internet today, users and applications must often make decisions based on the performance they expect to receive from other Internet hosts. For example, users can often view many Web pages in low-bandwidth or high-bandwidth versions, while other pages present users with long lists of mirror sites to chose from. Current techniques to perform these decisions are often ad hoc or poorly designed. The most common solution used today is to require the user to manually make decisions based on their own experience and whatever information is provided by the application. Previous efforts to automate this decision-making process have relied on isolated, active network probes from a host. Unfortunately, this method of making measurements has several problems. Active probing introduces unnecessary network traffic that can quickly become a significant part of the total traffic handled by busy Web servers. Probing from a single host results in less accurate information and more redundant network probes than a system that shares information with nearby hosts. In this paper, we propose a system called SPAND (Shared Passive Network Performance Discovery) that determines network characteristics by making shared, passive measurements from a collection of hosts. In this paper, we show why using passive measurements from a collection of hosts has advantages over using active measurements from a single host. We also show that sharing measurements can significantly increase the accuracy and timeliness of predictions. In addition, we present a initial prototype design of SPAND, the current implementation status of our system, and initial performance results that show the potential benefits of SPAND.

As the speed and the dynamic range of computer networks evolve, the issue of efficient traffic management becomes increasingly important. This work describes an approach to traffic management using explicit rate information provided to the source by the network. We present an asynchronous distributed algorithm for optimal rate calculation across the network, where optimality is understood in the maxmin sense. The algorithm quickly converges to the optimal rates and is shown to be well-behaved in transience. 1 Introduction 1.1 Background In the past decade several mechanisms for congestion control have been developed and implemented. DECbit [34] and Slow Start [20] are perhaps the best known. Both of these schemes were developed for connectionless networks with window flow control, in which the routers had no knowledge of the individual flows and their demands, the routes changed frequently and the header space was scarce. With the rapid increase of the ratio of propagation delay to...

This paper studies a simple strategy, proposed independently by Gallager [1] and Katevenis [2], for fairly allocating link capacity in a point-to-point packet network with virtual circuit routing. Each link offers its packet transmission slots to its user sessions by polling them in round-robin order. In addition, window flow control is used to prevent excessive packet queues at the network nodes. As the window size increases, the session throughput rates are shown to approach limits that are perfectly fair in the max-min sense. That is, the smallest session rate in the network is as large as possible and, subject to that constraint, the second-smallest session rate is as large as possible, etc. If each session has periodic input (perhaps with jitter) or has such heavy demand that packets are always waiting to enter the network, then a finite window size suffices to produce perfectly fair throughput rates. The round-robin method is considerably simpler than earlier strategies for achieving global fairness. The fair session rates are not explicitly computed, and the only overhead communication is that required for the window acknowledgments. The main drawback is that large windows are needed to achieve even approximately fair throughputs in some (hopefully rare) situations, and large windows permit large cross-network delays. Fortunately, the round-robin method offers other throughput guarantees that, while falling short of perfect fairness, do apply even for sessions with small windows. Such sessions are promised reasonable bounds on their cross-network packet delay as well.

This paper presents the packet-pair rate-based feedback flow control scheme. This scheme is designed for networks where individual connections do not reserve bandwidth and for the available bitrate (best-effort) component of integrated networks. We assume a round-robin-like queue service discipline in the output queues of the network's switches, and propose a linear stochastic model for a single conversation in a network of such switches. These model motivates the Packet-Pair rate probing technique, which forms the basis for provably stable discrete and continuous time rate-based flow control schemes. We present a novel state estimation scheme based on fuzzy logic. We then address several practical concerns: dealing with system startup, retransmission and timeout strategy, and dynamic setpoint probing. We present a finite state machine as well as source code for a model implementation. The dynamics of a single source, the interactions of multiple sources, and the behavior of packet-pai...

We consider a flow control mechanism that dynamically regulates the rate of data flow into a network based on feedback information about the network state. Such mechanisms have been introduced recently in a variety of networks including the Internet, and have been advocated for future high-speed networks. We first model the flow control mechanism by a discrete-space stochastic process and define appropriate performance measures for transient and steady-state regimes. However, the model does not appear to be analytically tractable and we study it through simulation. We then simplify it to a continuous-space deterministic (or fluid) model for which we can easily derive closed-form solutions. We find the analytical results for the fluid model to agree well with the simulation results obtained using the discrete-space model. Both models explicitly consider delay of the feedback information, thus making them relevant for high-speed networks. 1 Introduction In a computer network, packets g...

The Internet is expected to support various services, including best-effort services and guaranteed services. For best-effort services, we propose a new approach to achieving type-of-service (TOS) classes with adaptive nexthop routing. We consider two TOS classes, namely, delaysensitive and throughput-sensitive. As in routing protocols such as OSPF and integrated IS-IS, each node has a different next-hop for each destination and TOS class. Traditionally, a node has a single FCFS queue for each outgoing link, and the next-hops are computed using link measurements. In our approach, we attempt to isolate the two traffic classes by using for each outgoing link a separate FCFS queue for each TOS class; the link is shared cyclicly between its TOS queues. The next-hops for the delay-sensitive traffic adapts to link delays of that traffic. The next-hops for the throughput-sensitive traffic adapts to overall link utilizations. We compare our approach with the traditional approach using discret...

There has been considerable interest recently on flow control mechanisms where the data flow into a network is regulated based on feedback from the network. The window mechanism developed by Jacobson, in which sources use packet losses to adjust their window sizes, resulted in dramatic reduction of congestion in the Internet and has become an Internet standard. Recent studies have shown that Jacobsoh's algorithm gives rise to large-amplitude oscillations of window sizes and packet delays, which make it unsuitable to support applications such as packet video. In this paper, we describe and evaluate the performance of a flow control mechanism in which sources use packet round trip delays to adjust their window sizes. The objective is to maintain packet delays at a target level close to the minimum possible delay, and yet not severely restrict throughput. Simulation and experimental results show that the mechanism reduces the oscillations of window sizes, and provides connections with low average delay and low delay jitter, thus making it suitable to support applications such as a videoconference application currently under study, provided that some reservation mechanism is used in the intermediate gateways.

Future computer networks are expected to provide different types of service. For this purpose, new algorithms and protocols have been proposed for gateway scheduling, flow control, and routing. The interaction between these three components is crucial to the performance of the network. Existing work has studied only the interaction between scheduling and flow control, assuming static routing. In this paper, we investigate the interaction between scheduling and adaptive routing. We view the network as a dynamical system. We apply the Liapunov direct method to derive stability conditions for the routes of different traffic classes. We show how with scheduling support for routing, the routes of the traffic classes can be isolated, thereby improving the overall network performance. Monte-carlo simulation results are also presented. Categories and Subject Descriptors: C.2.1 [Computer-Communication Networks]: Network Architecture and Design---packet networks; store and forward networks; C....

We propose a new approach to achieving type-of-service (TOS) with adaptive next-hop routing in wide-area networks such as the Internet. We consider two traffic classes, namely delay-sensitive and throughput-sensitive. In routing protocols such as OSPF and integrated IS-IS, each node has a different nexthop for each destination and TOS. Traditionally, each node has a single FCFS queue for each outgoing link, and the next-hops are computed using link measurements. In our approach, we attempt to isolate the two traffic classes by using two FCFS queues for each outgoing link, one for each TOS; the link is shared cyclicly between the two TOS queues. The next-hops for the delay-sensitive traffic adapts to link delays of that traffic. The next-hops for the throughputsensitive traffic adapts to overall link utilizations. We compare our approach with the traditional approach using discrete-event simulation and Liapunov analysis (for stability of routes). The proposed approach offers ...