QoS-aware Dynamic Fog Service Provisioning
share
Recent advances in the areas of Internet of Things (IoT), cloud computing and big data have attributed to the rise of growing number of complex and useful applications. As IoT becomes more prevalent in our daily life, more data-intensive, delay-sensitive, and real-time applications are expected to emerge. Ensuring Quality of Service (QoS) in terms of bandwidth and low-latency for these applications is essential, and fog computing has been seen as one of the main enablers for satisfying these QoS requirements. Fog puts compute, storage, and networking resources closer to the user. In this paper, we first introduce the Dynamic Fog Service Provisioning problem, which is the problem of dynamically deploying new fog services (applications) on fog nodes, or releasing the services that are already deployed on fog nodes, in order to meet QoS constraints. The problem is formulated as an INLP task, and two heuristics are proposed to solve it efficiently. Finally, the heuristics are evaluated using a simulation based on the real- world traffic traces, and mobile augmented reality as the IoT application.
READ FULL TEXT