Master Thesis Patrick Schäfer
A prototypical tool to discover architecture changes based on multiple monitoring data sources for a distributed system.
Distributed architectures become more popular from year to year because of their support of agile project management and horizontal scaling of modern IT-systems. On the other hand, distributed systems lead to non-transparent and complex infrastructure. To address these challenges, Architecture Discovery Systems were developed. They are specialized to detect applications and IT-systems or business activities and business processes.
This work addresses the development of a prototypical Architecture Discovery System, which considers technical entities as well as business entities. In contrast to established systems, it provides one architecture model for all of these entities including relations between them. The prototype is based on data processing of a Service Repository and distributed tracing.
The prototype shows, how one can process and merge the different data sources to generate a comprehensive architecture model. It furthermore shows how to extend the model by entities of the business field and how to establish relations between the model's entities.
The result of the prototype is a comprehensive architecture model, which is dynamically detected and provided to external systems by a public interface. Architecture changes are provided in real time by utilizing streaming technologies. The architecture model is visualized as an adjacency matrix, which offers an alternative way of outlining complex architectures, being capable to represent even complex dependencies and relations in a clear way.
| Attribute | Value |
|---|---|
| Title (de) | Eine prototypische Implementierung zur Erkennung von Architekturänderungen eines verteilten Systems basierend auf unterschiedlichen Monitoring Datenquellen. |
| Title (en) | A prototypical tool to discover architecture changes based on multiple monitoring data sources for a distributed system. |
| Project | Multi-Level Monitoring Visualization |
| Type | Master's Thesis |
| Status | completed |
| Student | Patrick Schäfer |
| Advisor | Dr. Martin Kleehaus |
| Supervisor | Prof. Dr. Florian Matthes |
| Start Date | 15.04.2017 |
| Sebis Contributor Agreement signed on | 28.03.2017 |
| Checklist filled | Yes |
| Submission date | 15.11.2017 |