Design and exploration of load balancers for fog computing using fuzzy logic

This research work presents an exploratory study to find answers for managing traffic and payloads in edge/fog zone of the networks. Initial deployments have shown that edge computing has helped in fine grain customization of services and have produced better experiences for the users. However, at the same time, it also invites challenges such as overload, underload and imbalance in utilization of resources such as bandwidth, time responses, throughput etc. The main reason is highly constrained environment and restricted hardware capabilities of fog devices. The conventional algorithms, that work on finding the best routing paths for distributing and processing tasks, do not suffice with respect to Quality of Service (QoS) parameters of fog zone as their computations become unstable. An alternative approach is considered by adding software defined components, that can help in load balancing and managing the payloads using link analysis at the inter-nodes of cloud and fog zone. In this research work, a fuzzy load balancer is devised using different levels of design and tuning of fuzzy controls. This fuzzy logic based algorithm has been implemented for conducting link analysis as interconnects for managing traffic. The analysis shows that 3-level design is energy efficient for load balancing in fog zone due to reduced number of intervals in fuzzy design, reduced overhead in provisioning and improved responsiveness. Higher levels (5-level and 7-level) lead to creation of redundant fuzzy rules and wastage of resources.