An approach to automatically detect problems in restructured deployment models based on formalizing architecture and design patterns

For the automated deployment of applications, technologies exist which can process topology-based deployment models that describes the application’s structure with its components and their relations. The topology-based deployment model of an application can be adapted for the deployment in different...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Software-intensive cyber-physical systems 2019-06, Vol.34 (2-3), p.85-97
Hauptverfasser: Saatkamp, Karoline, Breitenbücher, Uwe, Kopp, Oliver, Leymann, Frank
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:For the automated deployment of applications, technologies exist which can process topology-based deployment models that describes the application’s structure with its components and their relations. The topology-based deployment model of an application can be adapted for the deployment in different environments. However, the structural changes can lead to problems, which had not existed before and prevent a functional deployment. This includes security issues, communication restrictions, or incompatibilities. For example, a formerly over the internal network established insecure connection leads to security problems when using the public network after the adaptation. In order to solve problems in adapted deployment models, first the problems have to be detected. Unfortunately, detecting such problems is a highly non-trivial challenge that requires deep expertise about the involved technologies and the environment. In this paper, we present (1) an approach for detecting problems in deployment models using architecture and design patterns and (2) the automation of the detection process by formalizing the problem a pattern solves in a certain context. We validate the practical feasibility of our approach by a prototypical implementation for the automated problem detection in TOSCA topologies.
ISSN:2524-8510
2524-8529
DOI:10.1007/s00450-019-00397-7