Scalability and Fault Tolerance for Real-Time Edge Applications

The rising demand of real time applications like health care, video surveillance and social media trend analysis, along with advancement in IoT and edge computing, ergs to use the edge computing for these real time applications. It can result in latency reduction, saving network bandwidth and efficient utilization of computing resources available at edge of the network. The nature of edge computing is distributed by-default hence it can offer a bench of resources for the applications executing in the edge computing, making it easy to handle the scalability requirements. Fault tolerance is of great concern for the real time applications in general and making it more important while working with edge computing. As, failures can result in disrupt of a critical activity, that can be more dangerous for a user or the real time application. The devices in the edge network of an IoT system are usually deployed individually and normally separate from each other having different capabilities and requirements. The lack of redundancy plan during device deployment makes fault tolerance and scaling operations more important for the edge-computing environment. This chapter discusses in detail the related issues, including the causes of faults and strategies for mitigating faults by providing fault tolerance and scaling in the distributed edge environment.