Home | Legals | KIT

Diplom/Masterarbeit: Automated Extraction of Palladio Component Models from Running Enterprise Java Applications


To ensure that a software system meets its performance requirements during system operation, the ability to predict its performance under different configurations and workloads is highly valuable. To enable performance prediction we need an abstraction of the real system that incorporates performance-relevant data, i.e., a performance model.  Based on such a model, performance analysis can be carried out. For example, if one observes a growing customer workload and assumes a steady workload growth rate, a performance model can help to determine when the system would reach its saturation point.

Unfortunately, building a predictive performance model manually requires a lot of time and effort. The model must be designed to reflect the system structure and capture its performance-relevant aspects. In addition, model parameters like service resource demands or system configuration parameters have to be determined. Current performance analysis tools used in industry mostly focus on profiling and monitoring transaction response times and resource consumption. The tools often provide large amounts of low-level data while important information needed for building performance models is missing. Given the costs of building performance models, techniques for automatic extraction of models based on observation of the system at run-time are highly desirable.

Thesis Goals

The aim of this thesis is to develop and implement a method for automated extraction of performance models of enterprise systems based on monitoring data collected during operation. State-of-the art industrial monitoring tools should be used to provide an end-to-end solution for the extraction. The focus is on the Java Platform, Enterprise Edition (Java EE) infrastructure. As performance model, the Palladio Component Model (PCM) is chosen. In order to evaluate the applicability of the approach, a case study with a representative enterprise application is conducted.


  • Automated extraction of Palladio Component Model instances
  • Work with the current version of Oracle WebLogic Server
  • Work with state-of-the-art industrial monitoring tools
  • Work with state-of-the-art Java Enterprise SPEC benchmarks
  • Collaboration with Oracle


Betreuer: Samuel Kounev