The replication of the experiment conducted by Schnoor and Hasselbring Comparing Static and Dynamic Weighted Software Coupling Metrics can either be executed online or locally.
Click on the badge to access the replication study online.
- Download and unzip or clone this repository.
- Download the kieker monitoring logs and copy them into the following folders but keep the existing
kieker.mapfiles.
- Experiment 1 ->
/replication/experiment/1/traces - Experiment 2 ->
/replication/experiment/2/traces - Experiment 3 ->
/replication/experiment/3/traces - Experiment 4 ->
/replication/experiment/4/traces
- Download the Jira byte code and extract the archives into the folowing folders.
- Jira byte code v7.3.0 ->
/replication/experiment/1/system - Jira byte code v7.4.3 ->
/replication/experiment/2/system - Jira byte code v7.7.1 ->
/replication/experiment/3/system - Jira byte code v7.7.1 ->
/replication/experiment/4/system
- Execute the batch files
1-scan.batand2-aggregate-calls.batfor each experiment one after another. - Execute
3-start-neo4j-server.batand explore the created dependency graphs of the selected experiment at http://localhost:7474/. - Before you can execute the jupyter notebook
/jupyter/Replication of Comparing Coupling Metrics.ipynb, you have to set up a corresponding environment. You can for example download Anaconda and install it. Furthermore, make sure thatjupyter notebook,pandas, andpy2neoare installed in the environment.
- Richard Müller, Dirk Mahler, Christopher Klinkmüller: Experiences in Replicating an Experiment on Comparing Static and Dynamic Coupling Metrics, 12th Symposium on Software Performance, CEUR, Leipzig, Germany, 2021.
You can find a complete list of publications on our website.