We consider the problem of routing traffic to optimize the performance of a congested network. We are given a network, a rate of traffic between each pair of nodes, and a latency function for each edge specifying the time needed to traverse the edge given its congestion; the objective is to route traffic such that the sum of all travel times—the total latency—is minimized. In many settings, it may be expensive or impossible to regulate network traffic so as to implement an optimal assignment of routes. In the absence of regulation by some central authority, we assume that each network user routes its traffic on the minimum-latency path available to it, given the network congestion caused by the other users. In general such a “selfishly motivated ” assignment of traffic to paths will not minimize the total latency; hence, this lack of regulation carries the cost of decreased network performance. In this article, we quantify the degradation in network performance due to unregulated traffic. We prove that if the latency of each edge is a linear function of its congestion, then the total latency of the routes chosen by selfish network users is at most 4/3 times the minimum possible total latency (subject to the condition that all traffic must be routed). We also consider the more general setting in which edge latency functions are assumed only to be continuous and nondecreasing in the edge congestion. Here, the total

We study the degradation in network performance caused by the selfish behavior of noncooperative network users. We consider a model of selfish routing in which the latency experienced by network traffic on an edge of the network is a function of the edge congestion, and network users are assumed to selfishly route traffic on minimumlatency paths. The quality of a routing of traffic is measured by the sum of travel times, also called the total latency. The outcome of selfish routing—a Nash equilibrium—does not in general minimize the total latency; hence, selfish behavior carries the cost of decreased network performance. We quantify this degradation in network performance via the price of anarchy, the worst-possible ratio between the total latency of a Nash equilibrium and of an optimal routing of the traffic. We show the price of anarchy is determined only by the simplest of networks. Specifically, we prove that under weak hypotheses on the class of allowable edge latency functions, the worst-case ratio between the total latency of a Nash equilibrium and of a minimum-latency routing for any multicommodity flow network is achieved by a singlecommodity

Abstract — A recent trend in routing research is to avoid inefficiencies in network-level routing by allowing hosts to either choose routes themselves (e.g., source routing) or use overlay routing networks (e.g., Detour or RON). Such approaches result in selfish routing, because routing decisions are no longer based on system-wide criteria but are instead designed to optimize hostbased or overlay-based metrics. A series of theoretical results showing that selfish routing can result in suboptimal system behavior have cast doubts on this approach. In this paper, we use a game-theoretic approach to investigate the performance of selfish routing in Internet-like environments, using realistic topologies and traffic demands in our simulations. We show that in contrast to theoretical worst cases, selfish routing achieves close to optimal average latency in such environments. However, such performance benefit comes at the expense of significantly increased congestion on certain links. Moreover, the adaptive nature of selfish overlays can significantly reduce the effectiveness of traffic engineering by making network traffic less predictable.

We address the issue of how to provide basic quality of service (QoS) in optical burst-switched WDM networks with limited fiber delay lines (FDLs). Unlike existing buffer-based QoS schemes, the novel offset-time-based QoS scheme we study in this paper does not mandate any buffer for traffic isolation, but nevertheless can take advantage of FDLs to improve QoS. This makes the proposed QoS scheme suitable for the next generation Optical Internet. The offset times required for class isolation when making wavelength and FDL reservations are quantified, and the upper and lower bounds on the burst loss probability are analyzed. Simulations are also conducted to evaluate the QoS performance in terms of burst loss probability and queuing delay. We show that with limited FDLs, the offset-time-based QoS scheme can be very efficient in supporting basic QoS.

Dynamic power management schemes (also called policies) can be used to control the power consumption levels of electronic systems, by setting their components in different states, each characterized by a performance level and a power consumption. In this paper, we describe power-managed systems using a finite-state, stochastic model. Furthermore, we show that the fundamental problem of finding an optimal policy which maximizes the average performance level of a system, subject to a constraint on the power consumption, can be formulated as a stochastic optimization problem called policy optimization. Policy optimization can be solved exactly in polynomial time (in the number of states of the model). We implemented a policy optimization tool and tested the quality of the optimal policies on a realistic case study.

Interference avoidance has been shown to reduce total square correlation (TSC) for given ensembles of user signature waveforms (codewords) in a synchronous CDMA system. In all experiments we have conducted, sequential application of interference avoidance produces an optimal codeword set when starting from randomly chosen initial codewords. Here we provide the rst formal proof of convergence to optimal codeword ensembles for greedy interference avoidance algorithms augmented by a technique called \class warfare" whereby users which reside in more heavily loaded areas of the signal space purposely interfere with (attack) the reception of users in less crowded areas. Coordination of deliberate interference by a complete class of aggrieved user is also sometimes necessary. Such \attacks" and subsequent codeword adjustment by attacked users are shown to strictly decrease TSC. Along the way we also show, using linear algebra and a variant of stochastic ordering, equivalence between minimiz...

Efficiency and portability are usually conflicting objectives for cluster-based network servers that distribute the clients ’ requests across the cluster based on the actual content requested. Our work is based on the observation that this efficiency vs. portability tradeoff has not been discussed before in the literature. To fill this gap, in this paper we study this tradeoff in the context of an interesting class of content-based network servers, the locality-conscious servers, using modeling and experimentation. Our analytical model gauges the potential performance benefits of portable and non-portable localityconscious request distribution with respect to a traditional, locality-oblivious server, as a function of multiple parameters. Based on our experience with the model, we design and evaluate a portable, locality-conscious server. Experiments with our server, a nonportable server, and a traditional server validate and confirm our modeling results under several real workloads. Based on our modeling and experimental results, our main conclusion is that portability should be promoted in cluster-based network servers with low processor overhead communication, given its relatively low cost 15%) in terms of efficiency. For clusters with high processor overhead communication, efficiency should be the overriding concern, as the cost of portability can be very high (as high as 98 % on 32 nodes). We also conclude that user-level communication can be useful even for non-scientific applications such as network servers.

This paper studies the instability of multiclass queueing networks. We prove that if a fluid limit model of the queueing network is weakly unstable, then the queueing network is unstable in the sense that the total number of customers in the queueing network diverges to infinity with probability one as time t ! 1. Our result provides a converse to a recent result of Dai [2] which states that a queueing network is positive Harris recurrent if a corresponding fluid limit model is stable. Examples are provided to illustrate the usage of the result. AMS 1991 subject classification: Primary 60K25, 90B22; Secondary 60K20, 90B35. Key words and phrases: multiclass queueing networks, instability, transience, Harris positive recurrent, fluid approximation, fluid model. Running title: Instability of queueing networks First draft in June 1995 Revised in July 1995 Revised in February 1996 Annals of Applied Probability, Vol. 6, pp 751--757 (1996) 1 Introduction This paper studies the transience ...

The need for network processors capable of forwarding IP packets at OC-192 and higher data rates has been well established. At the same time, there is a growing need for complex tasks, like packet classification and differentiated services, to be performed by network processors. At OC-768 data rate, a network processor has 9 nanoseconds to process a minimum-size IP packet. Such ultra high-speed processing, involving complex memory-intensive tasks, can only be achieved by multi-CPU distributed memory systems, using very high performance on-chip communication architectures. In this paper, we propose a novel communication network architecture for 8-CPU distributedmemory systems that has the potential to deliver the throughput required in next generation routers. We then show that our communication architecture can easily scale to accommodate much greater number of network nodes. Our network architecture yields higher performance than the traditional bus and crossbar yet has low implementation cost. It is quite flexible and can be implemented in either packet or circuit switched mode. We will compare and contrast our proposed architecture with busses and crossbars using metrics such as throughput and physical layout cost. 1.

This paper surveys performance models for distributed and replicated database systems. Over the last 20 years a variety of such performance models have been developed and they differ in (1) which aspects of a real system are or are not captured in the model (e.g. replication, communication, non-uniform data access, etc.) and (2) how these aspects are modeled. We classify the different alternatives and modeling assumptions, and discuss their interdependencies and expressiveness for the representation of distributed databases. This leads to set of building blocks for analytical performance models. To illustrate the work that is surveyed, we select a combination of these proven modeling concepts and give an example how to compose a balanced analytical model of a replicated database. We use this example to show how to derive meaningful performance values and to discuss the applicability and expressiveness of performance models for distributed and replicated databases. Finally, we compare the analytical results to measurements in a distributed database system.