Abstract Decentralized and unstructured peer-to-peer networks such as Gnutella are attractive for certain applicationsbecause they require no centralized directories and no precise control over network topologies and data placement. However, the flooding-based query algorithm used in Gnutella does not scale; each individual query gener-ates a large amount of traffic and, as it grows, the system quickly becomes overwhelmed with the query-induced load. This paper explores, through simulation, various alternatives to gnutella's query algorithm, data replicationmethod, and network topology. We propose a query algorithm based on multiple random walks that resolves queries almost as quickly as gnutella's flooding method while reducing the network traffic by two orders of mag-nitude in many cases. We also present a distributed replication strategy that yields close-to-optimal performance. Finally, we find that among the various network topologies we consider, uniform random graphs yield the bestperformance. 1 Introduction The computer science community has become accustomed to the Internet's continuing rapid growth, but even tosuch jaded observers the explosive increase in Peer-to-Peer (P2P) network usage has been astounding. Within a few months of Napster's [12] introduction in 1999 the system had spread widely, and recent measurement data suggeststhat P2P applications are having a very significant and rapidly growing impact on Internet traffic [11, 15]. Therefore, it is important to study the performance and scalability of these P2P networks. Currently, there are several different architectures for P2P networks:

Abstract—Open Shortest Path First (OSPF) is the most commonly used intra-domain internet routing protocol. Traffic flow is routed along shortest paths, splitting flow at nodes where several outgoing links are on shortest paths to the destination. The weights of the links, and thereby the shortest path routes, can be changed by the network operator. The weights could be set proportional to their physical distances, but often the main goal is to avoid congestion, i.e. overloading of links, and the standard heuristic recommended by Cisco is to make the weight of a link inversely proportional to its capacity. Our starting point was a proposed AT&T WorldNet backbone with demands projected from previous measurements. The desire was to optimize the weight setting based on the projected demands. We showed that optimizing the weight settings for a given set of demands is NP-hard, so we resorted to a local search heuristic. Surprisingly it turned out that for the proposed AT&T WorldNet backbone, we found weight settings that performed

A system of techniques is presented for optimizing OSPF/IS-IS weights for intradomain routing in a changing world, the goal being to avoid overloaded links. We address predicted periodic changes in traffic as well as problems arising from link failures and emerging hot-spots.

Traffic engineering is aimed at distributing traffic so as to "optimize" a given performance criterion. The ability to carry out such an optimal distribution depends on both the routing protocol and the forwarding mechanisms in use in the network. In IP networks running the OSPF or IS-IS protocols, routing is over shortest paths, and forwarding mechanisms are constrained to distributing traffic uniformly over equal cost shortest paths. These constraints often make achieving an optimal distribution of traffic impossible. In this paper, we propose and evaluate an approach, based on manipulating the set of next hops for routing prefixes, that is capable of realizing near optimal traffic distribution without any change to existing routing protocols and forwarding mechanisms. In addition, we explore the tradeoff that exists between performance and the overhead associated with the additional configuration steps that our solution requires. The paper's contributions are in formulating and evaluating an approach to traffic engineering for existing IP networks that achieves performance levels comparable to that offered when deploying other forwarding technologies such as MPLS.

An active network allows applications to inject customized programs into network nodes. This enables faster protocol innovation by making it easier to deploy new network protocols, even over the wide area. In this paper, we argue that the ability to introduce active protocols offers important opportunities for end-to-end performance improvements of distributed applications. We begin by describing several active protocols that provide novel network services and discussing the potential impact of these kinds of services on end-to-end application performance. We then present and analyze the performance of an active networking protocol that uses caching within the network backbone to reduce load on both servers and backbone routers. Keywords: active networks, caching, distributed applications, networking protocols, performance. 1 Introduction Traditionally, the function of a network has been to deliver packets from one endpoint to another. Processing within the network has been limited l...

but often the main goal is to avoid congestion, i.e. overloading of links, and the standard heuristic recommended by Cisco (a major router vendor) is to make the weight of a link inversely proportional to its capacity. We study the problem of optimizing OSPF weights for a given a set of projected demands so as to avoid congestion. We show this problem is NP-hard and propose a local search heuristic to solve it. We also provide worst-case results about the performance of OSPF routing vs. an optimal multi-commodity flow routing. Our numerical experiments compare the results obtained with our local search heuristic to the optimal multi-commodity flow routing, as well as simple and commonly used heuristics for setting the weights. Experiments were done with a proposed nextgeneration AT&T WorldNet backbone as well as synthetic internetworks.

Alternate path routing has been well-explored in telecommunication networks as a means of decreasing the call blocking rate and increasing network utility. However, aside from some work applying these concepts to unicast flows, alternate path routing has received little attention in the Internet community. We describe and evaluate an architecture for alternate path routing for multicast flows. For path installation, we design a receiver-oriented alternate path protocol and prove that it reconfigures multicast trees without introducing loops. For path computation, we propose a scalable local search heuristic that allows receivers to find alternate paths using only partial network information. We use a simulation study to demonstrate the ability of local search to find alternate paths approximately as well as a link-state protocol, with much lower overhead. I.

This paper evaluates a congestion control mechanism for reliable multicast applications that uses a small dynamic set of group members, or representatives, to provide timely and accurate feedback on behalf of congested subtrees of a multicast distribution tree. Our algorithm does not need to compute round-trip time (RTT) from all receivers to the source, nor does it require knowledge of group membership or network topology. Through simulations, we evaluate our algorithm with and without TCP cross traffic. This study demonstrates that while our algorithm makes use of bandwidth when available, it responds to congestion in a timely fashion thereby limiting data drops. When competing for bandwidth with TCP traffic, our algorithm gracefully relinquishes bandwidth. While not as aggressive as TCP, it does grab a reasonable portion of the bandwidth yet keeps the overall multicast packet loss low.

This paper considers resource allocation in a network with mobile agents competing for computational priority. We formulate this problem as a multi-agent game with the players being agents purchasing service from a common server. We show that there exists a computable Nash equilibrium when agents have perfect information into the future. From our game, we build a market-based CPU allocation policy and a strategy with which an agent may plan its expenditures for a multi-hop itinerary. We simulate a network of hosts and agents using our strategy to show that our resourceallocation mechanism effectively prioritizes agents according to their endowments and that our planning algorithm handles network delay gracefully. 1.

There exist a wide variety of network design problems that require a traffic matrix as input in order to carry out performance evaluation. The research community has not had at its disposal any information about how to construct realistic traffic matrices. We introduce here the two basic problems that need to be addressed to construct such matrices. The first is that of synthetically generating traffic volume levels that obey spatial and temporal patterns as observed in realistic traffic matrices. The second is that of assigning a set of numbers (representing traffic levels) to particular node pairs in a given topology. This paper provides an in-depth discussion of the many issues that arise when addressing these problems. Our approach to the first problem is to extract statistical characteristics for such traffic from real data collected inside two large IP backbones. We dispel the myth that uniform distributions can be used to randomly generate numbers for populating a traffic matrix. Instead, we show that the lognormal distribution is better for this purpose as it describes well the mean rates of origin-destination flows. We provide estimates for the mean and variance properties of the traffic matrix flows from our datasets. We explain the second problem and discuss the notion of a traffic matrix being well-matched to a topology. We provide two initial solutions to this problem, one using an ILP formulation that incorporates simple and well formed constraints. Our second solution is a heuristic one that incorporates more challenging constraints coming from carrier practices used to design and evolve topologies.