Computer Laboratory Home Page Search A-Z Directory Help
University of Cambridge Home Computer Laboratory
8th July 2004
Computer Laboratory > Research > Systems Research Group > NetOS > Seminars > 8th July 2004

Network Coding for Large Scale Content Distribution

Pablo Rodriguez Rodriguez
Up until recently, content distribution solutions consisted on placing dedicated equipment at certain places inside or at the edge of the Internet. The best example of such solutions is Akamai, which runs several tens of thousands of servers all over the world. However, in recent years, a new paradigm for Content Distribution has emerged based on a fully distributed architecture where commodity PCs are used to form a cooperative network and share their resources (storage, CPU, bandwidth).
The best example of an end-system cooperative architecture is the BitTorrent system. BitTorrent became extremely popular as a way of delivering the Linux distributions. BitTorrent works in the background of a Web browser, assisting with the uploading and downloading of files.
We propose a new scheme for end-system content distribution of large files that is based on network coding. With network coding, each node of the distribution network is able to generate and transmit encoded blocks of information. This is particularly important in large unstructured overlay networks, where the nodes need to make decisions based on local information only.
We compare network coding to other schemes that transmit unencoded information (i.e. blocks of the original file) and, also, to schemes in which only the source is allowed to generate and transmit encoded packets (e.g. Digital Fountain). We demonstrate the benefits of network coding under several realistic settings, and show that it can significantly outperform existing solutions for end-system content distribution.