Keywords:
Astract: The Digital Transformation of our society requires IT infrastructures to be more agile, more adaptive, and more connected than ever. At the same time, the owners of such infrastructures are confronted with an increase in regulatory pressure (e.g., the GDPR). These developments put a lot of stress on IT management and governance. To enable IT management and governance to better deal with these challenges, we propose to digitally transform IT governance and management itself by using a Digital Twin based approach. In line with this, we aim to create on ontology-driven Digital Twin for Governed IT Management (DT4GITM) framework. The goal of this framework is to act as a reference architecture for a Digital Twin based infrastructure that connects three interrelated systems: the IT governance processes, the governed IT management processes, and the managed organizational IT assets. The core of the framework involves a generic Governed IT Management (GITM) Domain Ontology, which is planned to be operationalized by a Knowledge Graph based approach that realizes an integrated view on the heterogenous data streams originating from the IT governance and management processes, and the managed IT assets. In this paper, we start by outlining the planned DT4GITM framework and the pivotal role of the GITM Domain Ontology within this. We then elaborate our incremental, and scenario- and case-driven strategy towards the development of the framework as a whole, and the GITM Domain Ontology in particular. This is followed by the elaboration of a specific DT4GITM scenario which serves as a first proof of principle.

Keywords:

Keywords:
Astract: A key challenge for elastic business processes is the resource-efficient scheduling of cloud resources in such a way that Quality-of-Service levels are met. So far, this has been difficult, since existing approaches use a coarse-granular resource allocation based on virtual machines. In this paper, we present a technique that provides fine-granular resource scheduling for elastic processes based on containers. In order to address the increased complexity of the respective scheduling problem, we develop a novel technique called GeCo based on genetic algorithms. Our evaluation demonstrates that in comparison to a baseline that follows an ad hoc approach a cost saving between 32.90% and 47.45% is achieved by GeCo while considering a high service level.

Keywords:
Astract: Research in areas of Internet of Things (IoT), Industry 4.0 (I4.0), Cyber Physical Systems (CPS), consider system-of-systems composed of hardware and software systems that seamless work together. Interoperability ranges from compatibility (systems do not disturb other systems, but no interaction between systems) to integration (systems that share a common model/worldview). Between these extremes a continuum of interoperability exists. In any case, interoperability is a model-driven approach - in contrast to data-driven or machine-learning approaches, for example. In general, models contain modules, connections, and behavior descriptions of different types. Integration and interoperability may not only be seen as (goal) states but also as processes. Changes in one system might trigger adaptation in other systems. However, in loose integration and interoperability settings there is more support for the evolution of models.