An Autonomic Strategy for Autoscaling in Smart City Platform Infrastructures
Smart city application development platforms receive, store, process, and display large volumes of data from different sources and have several users, such as citizens, visitors, government, and companies. The underlying computing infrastructure to support these platforms must deal with the highly dynamic workload of the different applications, with simultaneous access from multiple users and sometimes working with many interconnected devices. Such an infrastructure typically encompasses cloud platforms for data storage and
computation, capable of scaling up or down according to the demands of applications. This thesis proposes an autonomic approach for autoscaling smart city platform infrastructures. The approach follows the MAPE-K control loop to dynamically adjust the infrastructure in response to workload changes. It supports scenarios where the number of processing requests is unknown a priori. The performance of the approach has been evaluated upon the computational environment that supports Smart Geo Layers (SGeoL), a platform for developing real-world smart city applications.