ITR: Scalable, Attack-Resistant Peer-to-Peer Networks

点击翻译按钮获取中文摘要

Proposal Number: 0313160Title: "ITR: Scalable, Attack-Resistant Peer-to-peer Networks"PI: Jared SaiaAbstract:This project investigates the design of peer-to-peer networks which areboth provably scalable and attack-resistant. The networks arescalable in the sense that time, space and bandwidth resource costsfor all major operations are polylogarithmic in the number of nodes inthe network. The networks are attack-resistant in the sense that mostof the services are guaranteed to most of the participants, even aftermassive adversarial attack. The approach of this project is primarilytheoretical, i.e. algorithm design and analysis. However, the projectalso includes building prototype systems to empirically test thealgorithms.This project has potential for significant impact in other areas ofresearch. One major deliverable of the project is algorithms forcreating a p2p network which, even under attack, will enable sharingof massive data sets. This type of network could have an impact onseveral scientific disciplines, including, physics (e.g. for projectslike GriPhyN), astronomy (e.g. for projects like Sloan Digital SkySurvey), and biology (e.g. for projects like the Human GenomeProject). Another major deliverable of this project is algorithms forcreating a p2p network which, even under attack, will enabledistributed computation. This type of network would allow scientificprojects to tap into idle desktop computers to efficiently androbustly do lengthy computations. Astronomy (e.g. for projects likeSETI@home) and biology (e.g. for projects like FOLDING@home) areexamples of disciplines that would benefit from such a network.Finally, it is expected that the project will provide theoreticalinsights into the problem of building robust embedded networks, whichshare many similarities with p2p networks.