A virtual world or networked virtual environment is a computer-generated space used as a metaphor for interaction. Entities, driven by users (avatars) or by computers (virtual objects), enter and leave the world, move from one virtual place to another, and interact in real-time. Shared virtual reality applications provide a similar perception of the same scene for any two entities. The system we are designing and building intends to be scalable to an unlimited number of users and accessible by any computer connected to the Internet. It does not make use of any server and is solely based on a network of peers. The goal is to allow fast travelling (teleportation) inside the virtual world by considering the underlying network topology.
Peer-to-Peer systems have the interesting property of self-scaling,
which means that the amount of resources grows with the number of
participants. While there exist already a large number P2P systems for
file sharing, very little work has been done in the area of using P2P
systems for file backup. Typically, file back-up is done in a purely
centralized manner. Such an organization requires a large amount of
resources (disks, tape robot) and also some human intervention. On the
other hand there is an increasing number of PCs each equipped with a
local disk with a capacity of tens of Giga Bytes. The goal of our
research is to investigate how the local disks of a large number of
PCs can be organized in such a manner as to allow a highly reliable
file back-up system.
This research is supported by Microsoft Research Cambridge.
With the widespread availability of inexpensive broadband Internet
connections, a large number of bandwidth-intensive applications have now
become practical. This is the case of multimedia live streaming, for
which end-user's access connections once were the bottleneck. The
bottleneck is now mainly found on the server side, since the bandwidth
requirement at the server grows linearly with the size of the audience.
In the last few years research efforts have targeted the problem of
streaming content distribution, and mostly with a pure theoretical
approach. The goal of our research is to study and design a p2p live
streaming system that takes into account the inherent characteristics of
the Internet (uneven peer bandwidth distribution, common asymmetry
between inbound and outbound link capacities, position of the nodes in
the network) and of its users (variable willingness/ability to
cooperate) and exploits them to preserve the whole system's health and
to reward those peers who contribute.
This research is supported by France Telecom R+D.
We have analyzed BitTorrent, a very popular peer-to-peer application that allows distribution of very large contents to a large set of hosts. Our analysis of BitTorrent is based on measurements collected on a five months long period that involved thousands of peers. We assess the performance of the algorithms used in BitTorrent through several metrics. Our conclusions indicate that BitTorrent is a realistic and inexpensive alternative to the classical server-based content distribution.
Peer-to-peer networks have often been touted as the ultimate solution
to scalability. Cooperative content distribution is based on the
premise that the capacity of a network is as high as the sum of the
resources of its nodes: the more peers in the network, the higher its
aggregate bandwidth, and the better it can scale and serve new
peers. Such networks can thus spontaneously adapt to the demand by
taking advantage of available resources.
We evaluate the use of peer-to-peer networks for content distribution under various system assumptions, such as peer arrival rates, bandwidth capacities, cooperation strategies, or peer lifetimes. We argue that the self-scaling and self-organizing properties of cooperative networks pave the way for cost-effective, yet highly efficient and robust content distribution.
Two factors are crucial to the global effectiveness of the content distribution process: