In theory, multi homed Internet hosts, that is, hosts simultaneously connected to multiple Internet service providers (ISP) should see increased access capacity, be able to circumvent possible last-mile congestion problems, and experience improved end-to-end quality of service (QoS). In practice however, the advantages one can gain from multi homing are highly dependent on the path switching mechanism used, that is, on dynamically deciding which ISP should be used as a first-hop. This paper is a first step toward understanding the trade-off between performance improvements multi homing can help achieve and the complexity of the decisions that must be made. We measure changes in end-to-end network layer metrics (loss, latency, jitter) over the different paths available from a multi homed host to a large population of Internet hosts. Our measurements indicate that (1) in over 60% of the cases, one only needs to reevaluate the service provided by each ISP every minute to improve the performance of a specific metric, and that (2) in approximately 85% of the cases, decisions to switch from one ISP to another can be treated independently of the service metric of interest. We conclude that multi homing could in practice result in noticeable performance improvements.

Striping is a resource aggregation technique that can improve application performance for a multihomed host by utilizing multiple interfaces. In this position paper, we argue in favor of decoupling striping primitives from all of the traditional layers of the networking protocol stack, and instead resurrecting the session layer for striping. Session layer striping indeed allows applications to take advantage of multihoming, while avoiding most of the deployability issues traditionally linked with modiflying application layer code. We illustrate our argument with a strawman architecture for a session layer striping protocol, and sketch how such a protocol could be implemented.

We propose a service differentiated peer selection mechanism for peer-to-peer media streaming systems. The mechanism provides flexibility and choice in peer selection to the contributors of the system, resulting in high quality streaming sessions. Free-riders are given limited options in peer selection,if any, and hence receive low quality streaming. The proposed incentive mechanism follows the characteristics of rank-order tournaments theory that considers only the relative performance of the players, and the top prizes are awarded to the winners of the tournament. Using rank-order tournaments, we analyze the behavior of utility maximizing users. Through simulation and wide-area measurement studies, we verify that the proposed incentive mechanism can provide near optimal streaming quality to the cooperative users until the bottleneck shifts from the streaming sources to the network.

Residential multihoming provides path redundancy as well as path diversity that can be used to improve not just network availability, but also application-layer quality of service. The path diversity can be effectively utilized by a streaming application to achieve high quality streaming. We design a multipath multihoming media streaming system (MMS) that exploits the path diversity due to multihoming to improve the streaming quality. Through a series of experiments over the Internet, we show that MMS improves the quality of streaming sessions by more than 30% when connections are split over the available physical links and migrated from one path to another to avoid congested paths.

We design a tree-based forward digest protocol (TFDP) to verify data integrity in distributed media streaming for content distribution. Several challenges arise, including the timing constraint of streaming sessions, the involvement of multiple senders, and the untrustworthiness of these senders. A comprehensive comparison is presented on the performance of existing protocols and TFDP, with respect to communication and computation overhead. Both simulation and Internet-based experimental results are presented to demonstrate the effectiveness of TFDP.

We propose a rank-based peer-selection mechanism for peer-to-peer media streaming systems. The mechanism provides incentives for cooperation through service differentiation. Contributors to the system are rewarded with flexibility and choice in peer selection, resulting in high quality streaming sessions. Free-riders are given limited options in peer selection, if any, and hence receive low quality streaming. Through simulation and wide-area measurement studies, we verify that the mechanism can provide near optimal streaming quality to the cooperative users until the bottleneck shifts from the sources to the network.

We present the design, implementation, and evaluation of PROMISE, a novel peer-to-peer media streaming system encompassing the key functions of peer lookup, peer-based aggregated streaming, and dynamic adaptations to network and peer conditions. Particularly, PROMISE is based on a new application level P2P service called CollectCast. CollectCast performs three main functions: (1) inferring and leveraging the underlying network topology and performance information for the selection of senders; (2) monitoring the status of peers and connections and reacting to peer/connection failure or degradation with low overhead; (3) dynamically switching active senders and standby senders, so that the collective network performance out of the active senders remains satisfactory. Based on both real-world measurement and simulation, we evaluate the performance of PROMISE, and discuss lessons learned from our experience with respect to the practicality and further optimization of PROMISE.

TCP connection throughput is inversely proportional to the connection round trip time (RTT). To mitigate TCP bias to short RTT connections, a differentiated services traffic conditioner can ensure connections with long RTTs do not starve when connections with short RTTs get all extra resources after achieving the target rates. Current proposals for RTT-aware conditioners work well for a small number of connections when most TCP connections are in the congestion avoidance phase. If there is a large number of TCP connections, however, connections time-out and go to slow start. We show that current RTT-aware conditioners over-protect long RTT flows and starve short RTT flows in this case. We design and evaluate a conditioner based on RTT as well as the retransmission time-out (RTO). The proposed RTT-RTO aware traffic conditioner works well for realistic situations with a large number of connections. Simulation results in a variety of situations confirm that the conditioner mitigates RTT bias.

We design and evaluate an adaptive traffic conditioner to improve application performance over the differentiated services assured forwarding behavior. The conditioner is adaptive because the marking algorithm changes based upon the current number of flows traversing through an edge router. If there are a small number of flows, the conditioner maintains and uses state information to intelligently protect critical TCP packets. On the other hand, if there are many flows going through the edge router, the conditioner only uses flow characteristics as indicated in the TCP packet headers to mark without requiring per flow state. Simulation results indicate that this adaptive conditioner improves throughput of data extensive applications like large FTP transfers, and achieves low packet delays and response times for Telnet and WWW traffic.