Master Thesis MSTR-2020-39

BibliographyMathony, Tobias: Deployment-technology-agnostic management of running applications.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Master Thesis No. 39 (2020).
69 pages, english.

In recent years, a plethora of technologies emerged to automate the deployment of applications, which is, if manually performed, a complex and error-prone process. Since deployment technologies heavily differ in their feature sets, mechanisms, modeling languages, and deployment models, committing to a technology may result in a lock-in. Further, enterprises often use multiple deployment technologies for their applications, each fitting the respective need. However, managing multiple applications deployed with different technologies is tedious. Due to the aforementioned reasons, it is desirable to provide a normalized representation of running applications, as well as to enable the management of applications regardless of the technology used to deploy them. To bridge this gap, EDMMi is introduced, a normalized model to represent running applications independently of their deployment technology. Further, this work proposes an approach to retrieve the instance information of running applications in an automated manner. A mapping between technology-specific instance data and EDMMi is then used to derive a normalized model of the retrieved application instance. To gain advantage from standardization, a further transformation from EDMMi to the TOSCA instance model is provided. Afterwards, the standardized TOSCA instance model is enriched with additional management functionalities that can be executed on the running application based on the existing Management Feature Enrichment approach. As a result, the concept presented in this work enables the enrichment of running applications with standards-based executable workflows for additional management functionality, regardless of the technology used to deploy them. With the transformation of technology-specific instance information to a standardized instance model, also the management of running applications in one single place is enabled, uncoupled from their deployment technologies. To prove the feasibility of the proposed concept, a prototypical implementation and an accompanying case study within the EDMM Transformation Framework and the OpenTOSCA ecosystem is provided.

Full text and
other links
Department(s)University of Stuttgart, Institute of Architecture of Application Systems
Superviser(s)Leymann, Prof. Frank; Harzenetter, Lukas; Wurster, Michael
Entry dateDecember 17, 2020
   Publ. Institute   Publ. Computer Science