Dynamic Scheduling DDoS Signature Detection
Multi-cloud Network Management Multi-cloud Route Characterization Unraveling the Cloud Peering Fabric
Optics-based DDoS Defenses DDoS Signature Detection Climate Change Risks to Networks Colocation View of Internet Routes
Multi-task Weak Supervision De-noising Delay Measurements Forecasting Network Streams Application Behavior Modeling
Metro Fiber Characterization IPv6 Alias Resolution Mapping Internet Interconnections IPv6 Topology Discovery Geography of Eyeball ASes
Traffic Localization Web Complexity
BONSAI
Predicting Mental Health Incentivized Reviews Coarse View of Graphs
PRIME Mesh vs Multiple Trees ISP-Friendly Contribution Awareness Dissecting Performance Overlay Monitoring
Unbiased Sampling Peer Dynamics (Churn) DHT Lookup Performance Tsunami, A DHT Botnet Shared Files Monitoring DHT Traffic Overlay Topology Impact on [AS] Underlay
Publisher Behavior Peer-Level Performance Assessing Scalability

P2P Mesh-based Live Streaming

PRIME: Peer-to-Peer Receiver-drIven MEsh-based Streaming

The success of swarming content delivery has motivated a new approach to live Peer-to-Peer (P2P) streaming that we call mesh-based streaming. In this approach, participating peers form a random mesh and incorporate swarming content delivery to stream live content. Despite the growing popularity of this approach, neither the design tradeoffs nor the basic performance bottlenecks in mesh-based P2P streaming are well understood. We designed PRIME, the first mesh-based P2P streaming for live content that effectively incorporates swarming content delivery. We identify two performance bottlenecks in a mesh-based P2P streaming, namely bandwidth bottleneck and content bottleneck. We derive proper peer connectivity to minimize bandwidth bottleneck as well as an efficient pattern of delivery for live content over a random mesh to minimize content bottleneck. We show that the pattern of delivery can be divided into diffusion and swarming phases and then identify proper packet scheduling algorithm at individual peers. Using ns simulations, we examine key characteristics, design tradeoffs and the relationship between main system parameters.

Publications

Team

  • Nazanin Magharei (UO)
  • Reza Rejaie (UO)

Funding

  • This material is based upon work supported by the National Science Foundation under NSF CAREER Award CNS-0448639. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

Mesh vs Multiple Trees

Existing approaches to P2P streaming can be divided into two general classes: (i) tree-based approaches use push-based content delivery over multiple tree-shaped overlays, and (ii) mesh-based approaches use swarming content delivery over a randomly connected mesh. Previous studies have often focused on a particular P2P streaming mechanism and no comparison between these two classes has been conducted. In this project, we compare and contrast the performance of representative protocols from each class using simulations. We identify the similarities and differences between these two approaches. Furthermore, we separately examine the behavior of content delivery and overlay construction mechanisms for both approaches in static and dynamic scenarios. Our results indicate that the mesh-based approach consistently exhibits a superior performance over the tree-based approach. We also show that the main factors attributing in the inferior performance of the tree-based approach are (i) the static mapping of content to a particular tree, and (ii) the placement of each peer as an internal node in one tree and as a leaf in all other trees.

Publications

  • Mesh or Multiple-Tree: A Comparative Study of Live P2P Streaming Approaches
    Nazanin Magharei, Reza Rejaie, Yang Guo
    Proceedings of IEEE INFOCOM, pp. 1424-1432, Anchorage, Alaska, May 2007
    [acceptance rate 17%]
    [PAPER] [PPT]

Team

  • Nazanin Magharei (UO)
  • Reza Rejaie (UO)
  • Yang Guo (Thomson Lab)

Funding

  • This material is based upon work supported by the NSF CAREER Award CNS-0448639. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

ISP-Friendly Swarm-based Peer-to-Peer Streaming

P2P streaming applications have been widely used for scalable streaming of live multimedia over the Internet. Their high bandwidth demand potentially can result in generating a significant amount of network traffic. P2P applications build a random overlay among peers and often ignore the underlying network topology. This imposes a significant amount of costly traffic on ISPs. For this reason, ISPs try to reduce the costly inter-ISP traffic by various means and thus, localized the P2P traffic. Localization can significantly degrades the performance of live P2P streaming applications. In a localized P2P overlay, there may not be adequate shortcuts throughout the varous parts of the overlay to ensure the diversity of content in each neighborhood. Thus, localization may results in decreasing the diversity of content in each peer’s neighborhood and stretching the shape of the random P2P overlay. We propose the first ISP-friendly Swarm-based P2P streaming mechanism with the goal of maintaining the P2P streaming application performance while reducing inter-ISPs traffic. In essence, we design an ISP-friendly block scheduling algorithm that considers the localized topology of the P2P overlay for smart block scheduling. Our evaluation results demonstrate that through the proposed ISP-friendly mechanism, the delivered quality to individual peers in a localized overlay is comparable to the peers’ quality in a random overlay.

Publications

  • ISP-friendly Live P2P Streaming
    Nazanin Magharei, Reza Rejaie, Ivica Rimac, Volker Hilt, Markus Hofmann
    IEEE/ACM Transactions on Networking, Volume 22, Number 1, pp. 244 - 256, May 2013
    [PAPER]

Team

  • Nazanin Magharei (UO)
  • Reza Rejaie (UO)

Contribution Awareness

Peer-to-Peer streaming overlays consist of peers with heterogeneous, asymmetric and limited bandwidth. A reasonable approach to maximize the delivered quality to individual peers in such an environment is to ensure that delivered quality to each peer is proportional to its contribution. In essence the delivered quality in a resource-constraint P2P streaming system should be contribution aware. The contribution-aware paradigm not only promotes the fairness among peers, but also serves as an effective means to encourage peers to contribute. Incorporating contribution-awareness into P2P streaming is challenging. In this project, the contribution awareness mechanism is incorporated into a mesh-based P2P live streaming, and its performance is extensively studied. Our mail goal is to deliver a persistent quality to individual peers proportional to their contribution while maximizing the utilization of the resources in the system. Through extensive evaluation we show that the contribution-awareness mechanism in a mesh-based P2P streaming is achievable, i.e. peers received quality is proportional to their contribution while resources in the system is fully utilized.

Publications

  • Incorporating Contribution-Awareness into Mesh-based Peer-to-Peer Streaming Services
    Nazanin Magharei, Reza Rejaie, Yang Guo
    Peer-to-Peer Networking and Applications, Volume 4, Issue 3, pp. 231-250, September 2011
    [PAPER]

Dissecting Performance

Mesh-based Peer-to-Peer streaming mechanisms incorporate swarming content delivery and thus are able to support scalable streaming of live content. Their key component is a packet scheduling scheme at each peer that determines pulled packet from neighbors while accommodating in-time arrival and diversity of delivered packets. Besides packet scheduling scheme, the overall performance of a mesh-based P2P streaming mechanism also depends on the availability of excess resources in the system. Recently proposed mesh-based P2P streaming mechanisms have been evaluated in a scenario-rich setting. Thus, neither their performance in a resource constraint scenario nor the separate effect of packet scheduling and available resources on their performance is known. In this project, we dissect the performance of mesh-based P2P streaming mechanism and investigate the effect of packet scheduling and available resource on their performance. We argue that the global pattern of content delivery primarily determines behavior of a mesh-based P2P streaming system. We present a new evaluation methodology that properly captures this pattern. Using our evaluation methodology, we examine the performance of representative scheduling schemes and the role of available resources. Our findings provide useful insights in design and evaluation of mesh-based P2P streaming mechanisms.

Publications

  • On Performance Evaluation of Swarm-based Live Peer-to-Peer Streaming Applications
    Reza Rejaie, Nazanin Magharei
    Springer Multimedia Systems, Volume 20, Issue 4, pp. 415 - 427, July 2014
    [PAPER]

Team

  • Nazanin Magharei (UO)
  • Reza Rejaie (UO)

Overlay Monitoring

In Swarm-based Peer-to-Peer Streamiing (SPS) mechanisms, participating peers form a randomly connected mesh over which they incorporate swarm-like content delivery. In practice, a subset of participating peers may form clusters in the overlay due to various reasons such as localization of connectivity within edge ISPs. Despite the commonly held assumptions, the appearance of such clusters could significantly degrade the delivered quality to participating peers in SPS mechanisms.

This project examines the effect of overlay clustering on the performance of SPS mechanisms for live content. Leveraging the notion of two-phase content delivery in SPS mechanisms, we illustrate the effect of overlay clustering on content delivery. We propose the Overlay Monitoring and Repair (OMR) mechanism as a distributed and scalable approach to maintain proper overlay connectivity in SPS mechanisms. The key idea is to use delivered quality to individual peers as an indication of poor connectivity from other regions of the overlay. OMR employs a probabilistic approach to ensure an adequate number of properly-positioned peers reacts to detected clustering in the overlay without any coordination. Reacting peers rewire a small number of carefully-selected connections in the overlay to significantly improve the performance of content delivery.

Publications