This pipeline is part of the Continuous Integration of Performance Models (CIPM) approach. It is responsible for incremental calibration, updating the usage model and the initial self-validation of the calibrated performance models. Moreover, this repository includes a source code for automatic evaluation of the calibtrated performance models
The steps to setup the calibration pipeline are found on wiki.
The related publication for this repository are the following:
-ICSA 2020: Manar Mazkatli, David Monschein, Johannes Grohmann, and Anne Koziolek; Incremental calibration of architectural performance models with parametric dependencies.
-Qudous 2018: Manar Mazkatli and Anne Koziolek; Continuous integration of performance model.
More publications on the CIPM approach are found here.
For evaluating the experements described in ICSA20 paper, you can follow the instructions of the evaluation automation platform.
- The evaluation of the experement on CoCoMe case study (BookSale Calibration) can be found here.
- The evaluation of the experements on TeaStore casestudy (train calibration, trainForRecommender calibration) can be found under this link.
A recent version of this pipeline with more features is found here.