[Back]

J. Schleicher:
"Engineering and Management of heterogenous Smart City Application Ecosystems";
Supervisor, Reviewer: S. Dustdar, A. Voigt, U. Zdun; Institute of Information Systems, Distributed Systems Group, 2017; oral examination: 2017-02-27.

The recent advent and rapid success of the smart city paradigm has led to its widespread adoption in cities and their supporting ecosystems around the globe. Spear headed by multiple international
research initiatives more and more vital aspects of cities are becoming smart. This opens up a vast array of new application possibilities, but also brings along several novel challenges. Various areas like infrastructure, industry, government and most importantly citizens, of a smart city generate large amounts of data and create sophisticated tangled interactions leading to ever increasing complexity. Stakeholders in the smart city domain not only face the challenges of managing these complex systems themselves, they also need to be able to make informed decisions based on the massive amounts of data smart cities generate. In order for stakeholders to stay on top of this emerging complexity, while still seeing the big picture in this dynamic environment, it is
essential to provide a holistic interdisciplinary view on the city. To enable stakeholders to build applications that enable such a view, they need to be able to focus on their respective area of expertise without the burden of dealing with underlying complexities that arise from the large scale nature of smart cities. This calls for sensible abstractions that hide the complexities of operating, managing, and running complex smart city applications in a similar way as today´s
mobile application ecosystems do.
In this thesis we present novel approaches for infrastructure, operations, and data management for enabling such smart city applications ecosystems. First, we present an approach for
infrastructure-agnostic artifact deployment that allows easy integration of heterogenous smart city infrastructures as well as the independent evolution of smart city applications and infrastructures.
This enables a broader integration of, as well as an easy migration between, infrastructures for smart city applications. To address the key challenge of data management, we present an approach for modeling and management of data sources in the smart city domain. Our approach
allows for efficient, distributed data access for applications and introduces a simple technologyagnostic description of data sources for stakeholders. This mechanism enables the exposure of relevant data sources not only for other stakeholders in the same city, but also in other smart cities around the globe significantly extending the application spectrum of smart city applications. In the context of operation management, we present solutions to enable and improve the operation as well as evolution of smart city applications, while respecting the complex security and compliance constraints. We introduce a service mobility approach that can enable the execution, as well as significantly improve the results of, distributed analytical environments. Additionally, we present a method for continuous evolution of container application deployments, capable of integrating security and compliance constraints. By doing so, we further enable the use of software containers for building smart city applications, leading to a substantial increase in flexibility for
developers and operations teams alike. We integrate these approaches into a comprehensive middleware toolkit, the Smart City Operating System (SCOS) that serves as the central element for a Smart City Application Ecosystem (SCALE).We discuss the URBEM Smart City Application as an example of such an smart city application utilizing SCOS to realize a holistic interdisciplinary decision support system for smart cities. Finally, we evaluate our approach in the context of multiple scenarios and show that the contributions of this thesis significantly improve infrastructure, operations and, data management in smart city application ecosystems.

Smart City / Cloud Computing / Operation / Infrastructure / Deployment / Datamanagement