Enhancing online communities with cycle-sharing for social networks

Abstract

The Internet has made it possible to exchange information more rapidly on a global scale. A natural succeeding step was the creation of Social Networks where anyone in the world can share their experiences, content, and current information, using only their Internet-enabled personal computer or mobile devices. Under this scope, there are many Social Networks such as Facebook, Orkut, and Youtube each one exporting its own APIs to interact with its users and groups databases. Studies done on Social Networks show that they follow some properties like the Small-World property. Meaning that traversing friendship relations, vast numbers of other users could be reached from each single user (e.g., Friends of Friends), even though users usually only interact (on a daily basis) with a restrict group of friends. Considering that these networks could be regarded as enablingpeer-to-peer information sharing (albeit mediated by a centrally controlled infrastructure), employing them for cycle-sharing should be a great improvement for global distributed computing, by allowing public-resource sharing among trusted users and within online virtual communities. Resources from these types of networks can be used to further advance studies in other areas which may be too computational intensive for using a single computer or a cluster, e.g., to process data mined from the various Social Networks. We describe the design, development, and resulting evaluation of a web-enabled platform, calledCSSN: Cycle-Sharing in Social Networks. The platform leverages a Social Network (Facebook) to perform discovery of computational resources, thus giving the possibility for any user to submit his own jobs for remote processing. Walls, messages, and comments in Facebook are used as the underlying transport for CSSN protocol messages, achieving full portability with existing Social Networks. Globally, CSSN gives the chance for common users to unleash the untapped computing power hidden in Social Networks, and exploit it using the cycle-sharing paradigm to speedup their own (or common) applications' execution.