Home  | Legals | Data Protection | Sitemap | KIT
Dominik Werle

M.Sc. Dominik Werle

Room: 334
Phone: +49 721 608-41609
Fax: +49 721 608-45990
dominik werleSap7∂kit edu

Institut für Programmstrukturen und Datenorganisation (IPD)

Jun.-Prof. Dr.-Ing. Anne Koziolek

Karlsruher Institut für Technologie (KIT)
Gebäude 50.34
Am Fasanengarten 5
76131 Karlsruhe

Research Abstract

In the future energy system, large amounts of energy status data have to be collected from a geographically distributed set of sources. It is, however, often unclear how design decisions regarding the software and hardware system (e.g., the dimensioning of hardware resources, which processing middleware is used, or how often and in which granularity sensors provide data) influence the quality of the system (e.g., response time, cost or energy consumption). Trial-and-error is often not feasible because of availability requirements and costs. These decisions have to be made both when planning a system and when evolving the system. In the planning case, there often is no implementation or environment of the system which could be tested regarding the quality impact. In the evolution case, the change is often one that is not covered by historical data and can hence not be inferred using, say, a regression model. Both cases are supported using software quality models. Current modeling languages for this purpose are, however, often limited in their application to data-intensive software systems. They do not allow the explicit modeling of middleware components with quality impact (such as message queues or processing frameworks) and do not allow the description of data-dependent system behavior. In my thesis, the simulation of software performance for such systems is advanced. I will use the software architecture of a cluster that handles energy status data from an electronics packaging lab and of the planned Energy Lab 2.0 data processing infrastructure as two initial case studies. Based on these and from literature, I will derive recurring design questions and a suitable modeling language that includes data streams as first-class entities. The analysis of the resulting models is initially focused on performance. Other possible quality dimensions are for example the cost of the composed system, consistency guarantees made by messaging components or the reliability of the system.

Research interests

  • Energy-efficient software architectures
  • Software architectures for Big Data applications
  • Simulation and analysis of software quality
  • Model-driven and view-based software development (Vitruvius)

Open theses and student researcher jobs

  • If you are interested in a Bachelor's or Master's thesis topic in the context of software engineering and energy management systems, please do not hesitate to contact me.



A collection of software engineering challenges for big data system development.
Hummel, O.; Eichelberger, H.; Giloj, A.; Werle, D.; Schmid, K.
2018. Euromicro Conference on Software Engineering and Advanced Applications 2018 (SEAA 2018), Prague, CZ, August 29-31, 2018. Ed.: T. Bures, 362–369, Institute of Electrical and Electronics Engineers Inc. doi:10.1109/SEAA.2018.00066
HIPE – An energy-Status-Data set from industrial production.
Bischof, S.; Trittenbach, H.; Vollmer, M.; Werle, D.; Blank, T.; Böhm, K.
2018. 9th ACM International Conference on Future Energy Systems, e-Energy 2018; Karlsruhe; Germany; 12 June 2018 through 15 June 2018, 599–603, ACM, New York (NY). doi:10.1145/3208903.3210278
The palladio-bench for modeling and simulating software architectures.
Heinrich, R.; Werle, D.; Klare, H.; Reussner, R.; Kramer, M.; Becker, S.; Happe, J.; Koziolek, H.; Krogmann, K.
2018. 40th ACM/IEEE International Conference on Software Engineering, ICSE 2018; Gothenburg; Sweden; 27 May 2018 through 3 June 2018, 37–40, ACM, New York (NY). doi:10.1145/3183440.3183474
HIPE -- An Energy-Status-Data Set from Industrial Production.
Bischof, S.; Trittenbach, H.; Vollmer, M.; Werle, D.; Blank, T.; Böhm, K.
2018. International Workshop on Energy Data and Analytics (EDA 2018), Karlsruhe, June 12, 2018
Deriving Power Models for Architecture-Level Energy Efficiency Analyses.
Stier, C.; Werle, D.; Koziolek, A.
2017. Computer Performance Engineering : Proceedings of the 14th European Workshop, EPEW 2017, Berlin, Germany, 7th - 8th September 2017. Ed.: P. Reinecke, 214–229, Springer, Cham. doi:10.1007/978-3-319-66583-2_14
A Domain-Specific Language for Model Consistency.
Werle, D.
2016. Studierendenprogramm der Fachtagung „Modellierung 2016“. Hrsg.: R. Heinrich, Karlsruhe
Designing information hiding modularity for model transformation languages.
Rentschler, A.; Werle, D.; Noorshams, Q.; Happe, L.; Reussner, R.
2014. 13th International Conference on Modularity, Lugano, Switzerland, April 22-25, 2014, 217–228, ACM, New York (NY). doi:10.1145/2577080.2577094
Remodularizing Legacy Model Transformations with Automatic Clustering Techniques.
Rentschler, A.; Werle, D.; Noorshams, Q.; Happe, L.; Reussner, R.
2011. Workshop on Analysis of Model Transformations (AMT’14), Valencia, Spain, September 29, 2014. Ed.: J. Dingel, 4–13, CEUR-WS, Aachen