Testing scalability and performance of decentralized on-line social networks

Abstract

During the last years, a lot of research has been conducted in the field of Decentralized Online Social Networks (DOSNs). Due to limited dependency on a dedicated central-based infrastructure, such kind of social networking enhances the privacy of users' personal data and information in case of local attacks, yet on the other hand suffers in terms of scalability. In this paper, we propose a platform that enables the custom creation of a decentralized network configuration, along with all necessary modules for simulating the trafic of realworld online social networks and measuring in parallel the performance of several DOSN architectures across artificial data. The goal of our work is two fold since it i) verifies the benefits of the decentralized paradigm based on several conclusions drawn and ii) provides a platform that enables the provision of custom experimentation for researchers in the related field. © 2016 ACM.