Diplomarbeit DIP-3322

Wettinger, Johannes: Concepts for Integrating DevOps Methodologies with Model-Driven Cloud Management Based on TOSCA.
Universität Stuttgart, Fakultät Informatik, Elektrotechnik und Informationstechnik, Diplomarbeit Nr. 3322 (2013).
83 Seiten, englisch.
CR-Klassif.C.2.4 (Distributed Systems)
D.2.11 (Software Engineering Software Architectures)
D.2.12 (Software Engineering Interoperability)
K.1 (The Computer Industry)

The paradigm of Cloud computing introduces new approaches to manage IT services and applications. Those approaches overcome traditional IT infrastructure and service management. One of the main goals of Cloud computing is to automate the whole management of IT services in order to reduce costs and to make the execution of management tasks less error-prone. To make this happen, Cloud providers offer proprietary tools to create and manage services in the Cloud. However, when services get more complex it is hard to manage them because those tools aim to be simple and thus provide limited functionality only. In addition, a particular service that was built based on a certain Cloud offering is bound to this offering including all its management aspects. Consequently, a service cannot be easily moved from one Cloud provider's infrastructure to another one's infrastructure.

Today, tools and frameworks implementing so called "DevOps methodologies" can be used to realize management of Cloud services without binding a service to a particular Cloud provider. Nevertheless, complex services are still hard to manage by following those methodologies. To make such services manageable and enable automation of management tasks, a holistic service model is needed. Thus, model-driven Cloud management is an emerging paradigm to realize a holistic management approach for services in the Cloud. Because the DevOps approach and the model-driven approach are originating in different backgrounds, model-driven Cloud management does not cover some aspects of DevOps methodologies that are key for Cloud services.

This thesis is focused on integrating DevOps methodologies with model-driven Cloud management. The goal is to combine the strengths of both approaches in order to minimize the shortcomings of the individual approaches.

Volltext und
andere Links
PDF (1736045 Bytes)
Abteilung(en)Universität Stuttgart, Institut für Architektur von Anwendungssystemen
BetreuerSchwertle, Isabell; Binz, Tobias
Eingabedatum5. Dezember 2013
   Publ. Institut   Publ. Informatik