Abstract — Recently, overlay networks have emerged as a means to enhance end-to-end application performance and availability. Overlay networks attempt to leverage the inherent redundancy of the Internet’s underlying routing infrastructure to detour packets along an alternate path when the given primary path becomes unavailable or suffers from congestion. However, the effectiveness of these overlay networks depends on the natural diversity of overlay paths between two endhosts in terms of physical links, routing infrastructure, administrative control, and geographical distribution. Several recent studies realized that a measurable number of path outages were unavoidable even with use of such overlay networks. This stems from the fact that overlay paths might overlap with each other when overlay nodes are selected without considering the underlying topology. An overlay network’s ability to quickly recover from path outages and congestion is limited unless we ensure path independence at the IP layer. This paper proposes a novel framework for topologyaware overlay networks. In this framework, we expressly design overlay networks, aiming to maximize path independence without degrading performance. We develop measurement-based heuristics for 1) placement of overlay nodes inside an ISP and 2) selection of a set of ISPs. We base our analysis on extensive data collection from 232 points in 10 ISPs, and 100 PlanetLab nodes. On top of node placement, we present measurement-based verification to conclude that single-hop overlay routing performs as well as multi-hop routing with respect to both availability and performance. Our analysis results show that a single-hop overlay path provides the same degree of path diversity as the multihop overlay path for more than 90 % of source and destination pairs. Finally, we validate the proposed framework using real Internet outages to show that our architecture is able to provide a significant amount of resilience to real-world failures. I.

Dynamic overlay routing has been proposed as a way to enhance the reliability and performance of IP networks. The major premise is that overlay routing can bypass congestion, transient outages, or suboptimal paths, by forwarding traffic through one or more intermediate overlay nodes. In this paper, we perform an extensive simulation study to investigate the performance of dynamic overlay routing. In particular, we leverage recent work on available bandwidth (avail-bw) estimation, and focus on overlay routing that selects paths based on avail-bw measurements between adjacent overlay nodes. First, we compare two overlay routing algorithms, reactive and proactive, with shortest-path native routing. We show that reactive routing has significant benefits in terms of throughput and path stability, while proactive routing is better in providing flows with a larger safety margin (“headroom”), and propose a hybrid routing scheme that combines the best features of the previous two algorithms. We then examine the effect of several factors, including network load, traffic variability, link-state staleness, number of overlay hops, measurement errors, and native sharing effects. Some of our results are rather ∗ This work was supported by the NSF CAREER award ANIR-0347374, and by a Georgia Tech Broadband Institute (GTBI) grant. surprising. For instance, we show that a significant measurement error, even up to 100 % of the actual avail-bw value, has a negligible impact on the efficiency of overlay routing. 1

The Internet still lacks adequate support for QoS applications with real-time requirements. In great part, this is due to the fact that provisioning of end-to-end QoS to traffic that traverses multiple autonomous systems (ASs) requires a level of cooperation between ASs that is difficult to achieve in the current architecture. Recently, service overlay networks have been considered as an approach to QoS deployment that avoids these difficulties. In this study, we address the problem of the topological synthesis of a service overlay network, where endsystems and nodes of the overlay network (provider nodes) are connected through ISPs that supports bandwidth reservations. We express the topology design problem as an optimization problem. Even though the design problem is related to the (in general NP-hard) quadratic assignment problem, we are able to show that relatively simple heuristic algorithms can deliver results that are sometimes close to the optimal solution.

Abstract — Network proximity and latency estimation is an important component in discovering and locating services and applications. With the growing number of services and service providers in the large-scale Internet, accurately estimating network proximity/latency with minimal probing overhead becomes essential for scalable deployment. Although there exist a number of network distance estimation schemes, they either rely on extensive infrastructure support, require the IP address of the potential targets, falsely cluster distant nodes, or perform poorly with few measurement errors. We propose Netvigator, a scalable network proximity and latency estimation tool that uses information obtained from probing a small number of landmark nodes and intermediate routers (termed milestones) that are discovered en route to the landmarks, to identify the closest nodes. With very little additional probing overhead, Netvigator uses distance information to the milestones to accurately locate the closest nodes. We developed a Netvigator prototype and report our performance evaluation on Planet-Lab and in the intranet of a large enterprise. We also test its scalability using simulations. The performance of Netvigator is compared with Vivaldi and GNP. I.

Abstract not found

We propose a two-tier overlay multicast architecture (TOMA) to provide scalable and efficient multicast support for a variety of group communication applications. In TOMA, multicast service overlay network (MSON) is advocated as the backbone service domain, while end users in the access domains form a number of small clusters, in which an applicationlayer multicast protocol is used for the communication between the clustered end users. Our two-tier architecture is able to provide efficient resource utilization with less control overhead, especially for large-scale applications. It also alleviates the forwarding state scalability problem and simplifies multicast tree construction and maintenance when there are large numbers of groups ongoing in the networks. To help the MSON provider efficiently plan backbone service overlay, we provide several dimensioning algorithms to locate proxies, select overlay links, and allocate link bandwidth. Based on our architecture, we also suggest a cost-based pricing model for the overlay ISP to charge multicast groups. This pricing model would provide key incentives for both service providers and clients to adopt our proposed TOMA service. Extensive simulation studies are conducted and the results demonstrate that TOMA performs well in several common scenarios, it provides efficient multicast transmission comparable to IP multicast, and is scalable to group size as well as to the number of co-existing groups. We also run experiments and show that our dimensioning algorithms could efficiently plan the network resources with little penalty.

An application-aware Deterministic Overlay One-to-Many (DOOM) protocol is proposed for meeting heterogeneous QoS requirements of multiple end users of High-Bandwidth Sensor Actuator Network (HB-SAN) applications. Although DOOM is initially targeted for use in collaborative adaptive systems of weather radars, it has been designed for use in wider class of sensing systems. DOOM protocol performs rate-based application aware congestion control by selecting end user specific subset of the sensor data for transmission thus adapting to available network infrastructure under dynamic network conditions. Performance of DOOM is evaluated for radar networking using a combination of Planetlab as well as an emulation based test-bed. It is shown that DOOM protocol is able to meet individual end user QoS requirements as well as aggregate QoS requirements of different end users. Moreover, multiple DOOM streams are friendly to each other as well as to TCP cross- traffic sharing the bottleneck link.

Recently, a moderate amount of work has been reported on the use of overlay networks to support value-added network services, such as multicasting, Quality of Service (QoS), security, etc. To design an overlay network, the first step is to choose an overlay topology connecting all the overlay service nodes. When considering overlay topologies, several questions need to be answered first: How overlay topologies affect overlay routing performance? Which topologies can provide satisfactory performance? How can we construct efficient overlay topologies connecting all the overlay nodes? In this paper, we focus on the overlay network topology construction issue. First, we evaluate and compare the performance and overhead of the existing overlay topologies. Second, we formalize the overlay topology construction problem and propose two new heuristic methods to construct efficient overlay topologies. Simulation results have demonstrated the efficiency of the two proposed approaches. It is shown that overlay service performance varies significantly with respect to different overlay topologies. Thus, it is important to choose an appropriate overlay network topology. The knowledge of IP-layer topology information also benefits significantly in constructing efficient overlay topologies as inferred from the results.

multimedia content service delivery. Its efficiency is maximized when all the service recipients have identical needs. In reality however, the end users may have a heterogeneous set of requirements for different service levels as well as different service components, depending on their system and network capabilities. We propose the notion of Service Adaptive Multicast (SAM) that balances the tradeoffs between providing individualized service to each client and maintaining an efficient overlay multicast tree structure. The novel aspects of our approach are (a) the ability to augment and transform existing paths into service paths with the desired attributes; and (b) integration of two tree maintenance processes: a receiver-initiated just-in-time adaptation of the multicast service tree driven by application/user perceived QoS, and a demand-driven tree maintenance process geared towards long-term tree quality. We demonstrate the performance of our approach using simulations of large client population.

In real-time Internet based teleoperation systems, the operator controls the robot and receives feedback through the Internet. Supermedia refers to robotic control commands, video, audio, haptic feedback, and other sensory information in the control system. Traditional transport services may not be able to meet the timely transmission requirements and dynamic priority changes of supermedia streams. This paper aims to design an e#cient transport service for teleoperation applications.