%0 Journal Article %1 zhou2011 %A Zhou, Fangfei %A Zhang, Liang %A Franco, Eric %A Revis, Richard %A Mislove, Alan %A Sundaram, Ravi %D 2011 %K browser content_distribution_network p2p webserver %T WebCloud: Recruiting web browsers for content distribution %U http://arxiv.org/abs/1109.3791 %X We are at the beginning of a shift in how content is created and exchanged over the web. While content was previously created primarily by a small set of entities, today, individual users -- empowered by devices like digital cameras and services like online social networks -- are creating content that represents a significant fraction of Internet traffic. As a result, content today is increasingly generated and exchanged at the edge of the network. Unfortunately, the existing techniques and infrastructure that are still used to serve this content, such as centralized content distribution networks, are ill-suited for these new patterns of content exchange. In this paper, we take a first step towards addressing this situation by introducing WebCloud, a content distribution system for online social networking sites that works by re- purposing web browsers to help serve content. In other words, when a user browses content, WebCloud tries to fetch it from one of that user's friend's browsers, instead of from the social networking site. The result is a more direct exchange of content ; essentially, WebCloud leverages the spatial and temporal locality of interest between social network users. Because WebCloud is built using techniques already present in many web browsers, it can be applied today to many social networking sites. We demonstrate the practicality of WebCloud with microbenchmarks, simulations, and a prototype deployment.

@article{zhou2011, abstract = { We are at the beginning of a shift in how content is created and exchanged over the web. While content was previously created primarily by a small set of entities, today, individual users -- empowered by devices like digital cameras and services like online social networks -- are creating content that represents a significant fraction of Internet traffic. As a result, content today is increasingly generated and exchanged at the edge of the network. Unfortunately, the existing techniques and infrastructure that are still used to serve this content, such as centralized content distribution networks, are ill-suited for these new patterns of content exchange. In this paper, we take a first step towards addressing this situation by introducing WebCloud, a content distribution system for online social networking sites that works by re- purposing web browsers to help serve content. In other words, when a user browses content, WebCloud tries to fetch it from one of that user's friend's browsers, instead of from the social networking site. The result is a more direct exchange of content ; essentially, WebCloud leverages the spatial and temporal locality of interest between social network users. Because WebCloud is built using techniques already present in many web browsers, it can be applied today to many social networking sites. We demonstrate the practicality of WebCloud with microbenchmarks, simulations, and a prototype deployment. }, added-at = {2011-09-20T09:41:37.000+0200}, author = {Zhou, Fangfei and Zhang, Liang and Franco, Eric and Revis, Richard and Mislove, Alan and Sundaram, Ravi}, biburl = {https://www.bibsonomy.org/bibtex/2c2ef4585982f8c1be853834b6aabbbdd/schmidt2}, description = {WebCloud: Recruiting web browsers for content distribution}, interhash = {1021866f91906de0d87fa3ddd6adc823}, intrahash = {c2ef4585982f8c1be853834b6aabbbdd}, keywords = {browser content_distribution_network p2p webserver}, note = {cite arxiv:1109.3791Comment: 12 pages, under submission to conference}, timestamp = {2011-09-20T09:41:38.000+0200}, title = {WebCloud: Recruiting web browsers for content distribution}, url = {http://arxiv.org/abs/1109.3791}, year = 2011 }