No abstract

Abstract. In INterdomain Ingress Traffic Engineering (INITE), a "target" Autonomous System (AS) aims to control the ingress link at which the traffic of one or more upstream source networks enters that AS. In practice, ISPs often manipulate, mostly in a trial-and-error manner, the length of the AS-Path attribute of upstream routes through a simple technique known as prepending (or padding). In this paper, we focus on prepending and propose a polynomial-time algorithm (referred to as OPV) that determines the optimal padding for an advertised route at each ingress link of the target network. Specifically, given a set of "elephant" source networks and some maximum load constraints on the ingress links of the target AS, OPV determines the minimum padding at each ingress link so that the load constraints are met, when it is feasible to do so. OPV requires as input an AS-Path length estimate from each source network to each ingress link. We describe how to estimate this matrix, leveraging the BGP Looking Glass Servers. To deal with unavoidable inaccuracies in the AS-Path length estimates, and also to compensate for the generally unknown BGP tie-breaking process in upstream networks, we also develop a robust variation (RPV) of the OPV algorithm.

Abstract. Intelligent Route Control (IRC) technologies allow multihomed networks to dynamically select egress links based on performance measurements. TCP congestion control, on the other hand, dynamically adjusts the send-window of a connection based on the current path's available bandwidth. Little is known about the complex interactions between IRC and TCP congestion control. In this paper, we consider a simple dual-feedback model in which both controllers react to packet losses, either by switching to a better path (IRC) or by reducing the offered load (TCP congestion control). We first explain that the IRC-TCP interactions can be synergistic as long as IRC operates on larger timescales than TCP ("separation of timescales"). We then examine the impact of sudden RTT changes on TCP, the behavior of congestion control upon path changes, the effect of IRC measurement delays, and the conditions under which IRC is beneficial under two path impairment models: short-term outages and random packet losses.

Models for network topology form a crucial component in the analysis of protocols. This paper systematically investigates a variety of evolutionary models for autonomous system (AS) level Internet topology. Evolution-based models produce a topology incrementally, attempting to reflect the growth patterns of the actual topology. While evolutionary models are appealing, they have generally not agreed as closely with measurements of real data as non-evolutionary models. We attempt to understand what factor contributes to a "good" evolutionary model. Our systematic study consists of a relatively generic evolutionary model framework, which we populate with different choices for the components. This allows us to compare a variety of instances of models to measurements from real data sets. We study issues such as the initial topology, the type of preferential connectivity used when adding edges, and the role of "growth" edges added between existing nodes. We find that appropriate instantiation of the framework can provide topologies that agree closely with real data. We also use our work to highlight several crucial open problems in topology modeling. GLOBECOM