With network components increasingly reliable, routing is playing an ever greater role in determining network reliability. This has spurred much activity in improving routing stability and reaction to failures, and rekindled interest in centralized routing solutions, at least within a single routing domain. Centralizing decisions eliminates uncertainty and many inconsistencies, and offers added flexibility in computing routes that meet different criteria. However, it also introduces new challenges; especially in reacting to failures where centralization can increase latency. This paper leverages the flexibility afforded by centralized routing to address these challenges. Specifically, we explore when and how standby backup forwarding options can be activated, while waiting for an update from the centralized server after the failure of an individual component (link or node). We provide analytical insight into the feasibility of such backups as a function of network structure, and quantify their computational complexity. We also develop an efficient heuristic reconciling protectability and performance, and demonstrate its effectiveness in a broad range of scenarios. The results should facilitate deployments of centralized routing solutions. This material is posted here with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of the University of Pennsylvania's products or services. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to pubs-permissions@ieee.org. By choosing to view this document, you agree to all provisions of the copyright laws protecting it.This conference paper is available at ScholarlyCommons: http://repository.upenn.edu/ese_papers/521 On the Feasibility and Efficacy of Protection Routing in IP NetworksKin-Wah Kwong * , Lixin Gao † , Roch Guérin * , and Zhi-Li Zhang ‡ * University of Pennsylvania, † University of Massachusetts, ‡ University of Minnesota kkw@seas.upenn.edu, lgao@ecs.umass.edu, guerin@ee.upenn.edu, zhzhang@cs.umn.edu Abstract-With network components increasingly reliable, routing is playing an ever greater role in determining network reliability. This has spurred much activity in improving routing stability and reaction to failures, and rekindled interest in centralized routing solutions, at least within a single routing domain. Centralizing decisions eliminates uncertainty and many inconsistencies, and offers added flexibility in computing routes that meet different criteria. However, it also introduces new challenges; especially in reacting to failures where centralization can increase latency. This paper leverages the flexibility afforded by centralized routing to address these challenges. Specifically, we explore when and how standby backup forwarding options can be activated, while waiting for an update from the c...

Distributed routing algorithms may give rise to transient loops during path recomputation, which can pose significant stability problems in high-speed networks. We present a new algorithm, Distributed Path Computation with Intermediate Variables (DIV), which can be combined with any distributed routing algorithm to guarantee that the directed graph induced by the routing decisions remains acyclic at all times. The key contribution of DIV, besides its ability to operate with any routing algorithm, is an update mechanism using simple message exchanges between neighboring nodes that guarantees loop-freedom at all times. DIV provably outperforms existing loop-prevention algorithms in several key metrics such as frequency of synchronous updates and the ability to maintain paths during transitions. Simulation results quantifying these gains in the context of shortest path routing are presented. In addition, DIV's universal applicability is illustrated by studying its use with a routing that operates according to a non-shortest path objective. Specifically, the routing seeks robustness against failures by maximizing the number of next-hops available at each node for each destination. Abstract-Distributed routing algorithms may give rise to transient loops during path recomputation, which can pose significant stability problems in high-speed networks. We present a new algorithm, Distributed Path Computation with Intermediate Variables (DIV), which can be combined with any distributed routing algorithm to guarantee that the directed graph induced by the routing decisions remains acyclic at all times. The key contribution of DIV, besides its ability to operate with any routing algorithm, is an update mechanism using simple message exchanges between neighboring nodes that guarantees loopfreedom at all times. DIV provably outperforms existing loopprevention algorithms in several key metrics such as frequency of synchronous updates and the ability to maintain paths during transitions. Simulation results quantifying these gains in the context of shortest path routing are presented. In addition, DIV's universal applicability is illustrated by studying its use with a routing that operates according to a non-shortest path objective. Specifically, the routing seeks robustness against failures by maximizing the number of next-hops available at each node for each destination.

While the overall bandwidth of peer-to-peer live video streaming system scales automatically as peers collectively contribute the bandwidth, each peer also demands to download at the specified video playback rate so as to play the video smoothly. Therefore, a fundamental problem arisen is how to balance the bandwidth supply and demand in the peer-to-peer system to enjoy peers with the best video quality.To address this problem, we propose a fully distributed peer-topeer video streaming framework which automatically adapts the network towards full bandwidth utilization. Our design possesses two unique features. First, a special link-level homogenous overlay network is formed in which all the overlay links approach to have an identical bandwidth value. With such a feature, video flowing through the overlay links will not encounter any bottlenecks, and peers can thus achieve the guaranteed downloading rates. Second, based on the peer downloading rate observed locally at the streaming server, the server can adaptively adjust the video playback rate so that peers can achieve the best video quality with full bandwidth utilization. The effectiveness of our framework is verified through extensive simulations.

The sharing and dissemination of online content is one of the main purposes of social network applications, and the amount of content accessed through them, in particular through portable devices such as smartphones and PDAs, is expected to increase. Consumption of online content, however, does not require a continuous online presence. Content can be downloaded, consumed, modified, and uploaded at different times. An opportunity to improve a user's access to up-to-date information from its own social network is to take advantages of opportunistic contacts between mobile devices, \ie without waiting for connectivity to the network infrastructure. In other words, users of a social network application may receive more fresh content with no extra infrastructure deployment, simply by communicating with mobile devices of other users, in a delay-tolerant manner. Assessing the magnitude of this improvement is, however, challenging. For example, the frequency and patterns of such contacts are partly a function of the social connectivity of users, and so will be the availability of relevant information to share and more importantly the willingness to share that information. All these influence in non-trivial ways the gains that can be realized through opportunistic contacts. The paper's main contribution is in providing a quantitative handle through which these gains can be estimated, while accounting for the above factors. ABSTRACTContacts between mobile users provide opportunities for data updates that supplement infrastructure-based mechanisms. While the benefits of such opportunistic sharing are intuitive, quantifying the capacity increase they give rise to is challenging because both contact rates and contact graphs depend on the structure of the social networks users belong to. Furthermore, social connectivity influences not only users' interests, i.e., the content they own, but also their willingness to share data with others. All these factors can have a significant effect on the capacity gains achievable through opportunistic contacts. This paper's main contribution is in developing a tractable model for estimating such gains in a content update system, where content originates from a server along multiple channels, with blocks of information in each channel updated at a certain rate, and users differ in their contact graphs, interests, and willingness to share content, e.g., only to the members of their own social networks. We establish that the added capacity available to improve content consistency through opportunistic sharing can be obtained by solving a convex optimization problem. The resulting optimal policy is evaluated using traces reflecting contact graphs in different social settings and compared to heuristic policies. The evaluation demonstrates the capacity gains achievable through opportunistic sharing, and the impact on those gains of the structure of the underlying social network.

No abstract