Due to the limited computation and storage capacities, mobile and Internet of Things (IoT) devices rely on offloading. That is a task of sending data- and compute-intensive application components to the remote infrastructure for execution. However, offloading to the mainstream cloud architecture is not suitable for the real-time applications, as the cloud is not able to answer Quality of Service (QoS) requirements defined in Service Level Agreement (SLA) due to its own drawbacks: high latency, lack of mobility support and geo-distribution.
Therefore, we exploit edge computing resources distributed in user proximity as they provide: low latency, local network and traffic conditions awareness, location and mobility support. Moreover, we investigate novel strategies for monitoring and controlling QoS (response time, availability, responsiveness) in edge deployment, with a goal of non-intermittent service in case of user mobility in dynamic operating conditions such as smart city scenario.