Architecture-Level Software Performance Models for Online Performance Prediction

Proactive performance and resource management of modern IT infrastructures requires the ability to predict at run-time, how the performance of running services would be affected if the workload or the system changes. In this thesis, modeling and prediction facilities that enable online performance prediction during system operation are presented. Analyses about the impact of reconfigurations and workload trends can be conducted on the model level, without executing expensive performance tests.