(cache) The Coral Content Distribution Network

*"You make it fun; we'll make it run"* The Coral Content Distribution Network
	Home	••	Project home page
	Overview	••	Brief overview and news
	Usage	••	Coral Wiki and FAQ
	Lists	••	Mailing Lists
	Pubs	••	Publications and people
	Download	••	Plugins and source code
	Illuminati	••	Network measurement project
	OASIS	••	OASIS anycast service

Our Goal

Are you tired of clicking on some link from a web portal, only to find that the website is temporarily off-line because thousands or millions of other users are also trying to access it? Does your network have a really low-bandwidth connection, such that everyone, even accessing the same web pages, suffers from slow downloads? Have you ever run a website, only to find that suddenly you get hit with a spike of thousands of requests, overloading your server and possibly causing high monthly bills? If so, CoralCDN might be your free solution for these problems!

Using Coral

Taking advantage of CoralCDN is simple. Just append

.nyud.net

to the hostname of any URL, and your request for that URL is handled by Coral! Try this page, or any other site:

Current Deployment

260 servers world-wide

What is Coral?

Coral is a free peer-to-peer content distribution network, comprised of a world-wide network of web proxies and nameservers. It allows a user to run a web site that offers high performance and meets huge demand, all for the price of a $50/month cable modem.

Publishing through CoralCDN is as simple as appending a short string to the hostname of objects' URLs; a peer-to-peer DNS layer transparently redirects browsers to participating caching proxies, which in turn cooperate to minimize load on the source web server. Sites that run Coral automatically replicate content as a side effect of users accessing it, improving its availability. Using modern peer-to-peer indexing techniques, CoralCDN will efficiently find a cached object if it exists anywhere in the network, requiring that it use the origin server only to initially fetch the object once.

One of Coral's key goals is to avoid ever creating hot spots in its infrastructure. It achieves this through a novel indexing abstraction we introduce called a distributed sloppy hash table (DSHT), and it creates self-organizing clusters of nodes that fetch information from each other to avoid communicating with more distant or heavily-loaded servers.

A preliminary deployment of CoralCDN has been online since March 2004, running on the PlanetLab testbed. As of January 2006, it receives about 25 million requests per day from more than 1 million unique clients.

Check out some sites currently using CoralCDN.

(What's with the Google ads? We are generally interested in understanding how IP addresses and public information characterize Web clients (see our illuminati research project). One related question is how such information plays a role in pay-per-click advertising. So...to help us better understand how such systems work in practice, we decided to run some ourselves.)