This file specifies the hardware and software requirements that apply. Different software requirements apply, depending on whether you only want to repeat our experiments in a Docker container, or whether RaQuN is to be used as a library.
The following requirements apply for repeating the experiments presented in our paper. If you want to run the experiments on your own datasets, more or less resources might be required.
Any CPU should work. However, your CPU directly influences the time required to run the experiments, how many executions can
be run in parallel, and the measured runtimes. We ran the experiments on a Intel® Xeon® Processor E7-4880 at 2.90GHz.
Be aware that running the experiments in parallel will negatively affect the validity of the runtime evaluation, if your
processor does not have the necessary capabilities (number of (virtual) cores).
Some experiments (e.g., matching ArgoUML) might require large amounts of working memory. We advise to run the experiments on a system with at least 16GB RAM. If the experiments are to be run in parallel, more resources will be required, depending on the number of runs.
The artifacts, experimental results, and docker objects might require up to 10GB of storage. If the experiments are to be run in parallel, more resources will be required, depending on the number of runs.
The following software requirements apply, if you plan to run the experiments presented in our paper with Docker.
We provide batch scripts and bash scripts with parameterized Docker calls. You can use any OS supporting one of the script types. We tested the scripts on Windows 10, WSL2.0 with Ubuntu 20.04 LTS, and Manjaro Linux. If your OS supports neither batch nor bash scripts, you can execute the Docker calls in the scripts manually by copying them to your terminal.
To run our experiments in Docker, you only require Docker to be installed and running on your system. Please refer to the official documentation if you run into any problems when starting the Docker Daemon.
The following software requirements apply only if you plan to run the experiments presented in our paper without Docker.
We encourage you to use Docker though. It is easier to run, more robust, and
you can also quickly run the experiments with a new setup by rebuilding the Docker image.
RaQuN is written in Java and the result evaluation is written in Python-3. Therefore, you can run the experiments on any OS on which you can install the required software (see below).
The following requirements apply, if you want to run and evaluate the experiments without Docker:
- JDK-11
- Maven, which is also integrated in most Java IDEs
- Python-3.8
- matplotlib
- A tool for inflating archives (.zip). Examples are 7-ZIP or unzip. Windows 10 and most Linux distribution already have a built-in archival tool.