When the Internet protocols were standardized in the late 80s, most distributed applications were point-to-point, such as email, telnet, and ftp. Over the last few years, the nature of Internet traffic has changed dramatically. Driven by the production and distribution of vast amounts of user-created content, current Internet traffic is overwhelmingly many-to-many, leading to a mismatch between application demand and the underlying protocols.
This project is to develop a peer-to-peer solution, called OneSwarm, to serve as a universal communication layer for multi-point communication. By relying solely on software running on end-hosts, OneSwarm will avoid the high operational cost of infrastructure solutions and the deployability issues of clean-slate redesigns to the network architecture. Among the open questions that need answers: can swarming techniques work and meet desired levels of service quality when content files are no longer large, when there are strict deadlines on delivery, when the communication is not all-to-all, or when different nodes have different needs, all without working at cross-purposes to ISP objectives? This proposal will also investigate incentives to encourage end hosts to contribute resources to support demanding applications, by allowing users to trade resources across time, across swarms, and across applications.
Peer-to-peer systems offer lots of promise for scalable content distribution -- deployability, robustness, resistence to DoS attacks, and self-configurability. But to fully realize their potential, we need a generic platform -- one which provides a general-purpose communication layer to attain service predictability in end-host based systems and also one which shares incentives between applications and over time. Our research addresses these challenges and has yielded a number of widely-used systems. We examined the issue of incentive compatibility in BitTorrent and arrived at a surprising conclusion: although BitTorrent's mechanisms discourages free riding, the bulk of BitTorrent's performance has little to do with incentives, with the dominant performance effect in practice is altruistic contribution on the part of a small minority of high capacity peers. This observation led to the development of a more robust and high performing BitTorrent client called BitTyrant, which has been downloaded by more than a million users. OneSwarm, a file sharing system was designed to reduce the cost of privacy to the average user, since currently popular P2P file sharing networks expose users to silent, third party monitoring of their behavior. We developed novel techniques for efficient, robust, and privacy-preserving lookup and data transfer. We provided users flexible control over their privacy by defining sharing permissions and trust at the granularity of individual data objects and peers. The OneSwarm client is publicly available for download on Windows, Mac~OS~X, and Linux, and it is in widespread use around the globe. Being largely network-oblivious, many P2P applications may lead to inefficient network resource usage and/or low application performance. To address this, we proposed a simple architecture called P4P to allow for more effective cooperative traffic control between applications and network providers. We conducted extensive simulations and real-life experiments on the Internet to demonstrate the feasibility and effectiveness of P4P. Our experiments demonstrated that P4P either improves or maintains the same level of application performance of native P2P applications, while, at the same time, it substantially reduces network provider cost compared with either native or latency-based localized P2P applications. The technology has been field-tested by five major ISPs (Verizon, AT&T, Comcast, Telefonica, and Bell-Canada), and is currently under consideration to be ratified as an Internet standard by the IETF.
