Achieving Fairness through Load-Balancing in Social Cloud Computing Networks

Abstract

Cloud-based computing networks have taken over the digital landscape. From small non-profits to large multinational corporations, more and more entities have been offloading computing effort to the cloud in order to take advantage of the increased cost-efficiency and scalability of cloud computing. One of the new types of cloud that have emerged is the P2P cloud, which disengages from a traditional datacenter setup by allowing users to instead share their own computing hardware into a cloud to take advantage of cloud computing’s advantages at an even lower cost. However, this new paradigm comes with a slew of challenges, notably, security when operating with the devices of strangers and fairness when not operating in a market-based system. This paper aims to address these two issues by proposing an algorithm based on social credits for a P2P cloud system that uses a social network to establish its security measures. The project implements the Social Credit algorithm along with two other task-migration based load balancing algorithms adapted for a P2P social cloud in Cloudsim Plus, and the results are compared in terms of general metrics and fairness.