Over the past couple of years, the demand for computation resources has been increasing, due to the explosive growth of internet of things (IoT) devices in various application domains such as smart healthcare, smart city, intelligent transport, etc. The limited bandwidth and high latency existing in the traditional cloud computing system cannot meet user requirements. The data-intensive, latency-sensitive, latency-tolerant characteristics of multiple IoT applications pose major challenges to the current state-of-the-art centralized cloud architecture in terms of quality-of-service performance. To meet the computational requirements of various geo-distributed heterogeneous real-time IoT applications, this chapter proposes an edge computing paradigm to address the existing challenges in an ad-hoc manner, such that a better quality of service (QoS) may be experienced by the user. The resources in the edge layer are highly distributed, and hence we integrate the software-defined network (SDN) in the edge layer which primarily works to allocate various IoT tasks so that fewer IoT applications need to be offloaded to the cloud computing layer.