Case studies shared by users
Xavier Devroey Postdoc researcher SERG group of TUDelft
We are using the two BSR servers to execute large scale experiments with EvoCrahs with different configurations for the guided genetic algorithm. We estimate that until now we have consumed 2-3 weeks full time CPU time to set up the execution environment and run the experiments, and we will continue to use the infrastructure on regular basis. It allows us to have faster results (avoiding waiting for weeks if we had to use a regular machine) and scale to have relevant empirical data, which is very important for the validity of our research in search-based software testing.
Alifah syamsiyah PhD student at the AIS group of TU/e
I work on developing new process mining techniques that are able to deal with huge event logs recorded for processes executed in possibly highly variable and heterogeneous contexts. I use BSR infrastructure particularly in MySQL server to run my experiment in the database. In the experiment I extended the size of event data until it does not fit into my computer's memory. Then we utilize BSR server to store the data and now I can easily compute, analyze, and run my experiment.
Mauricio Aniche Postdoc researcher at the SERG group of TU Delft
I am applying code smells detection techniques to the entire change history of a very large sample of projects. These tools often take a few seconds per project. Thus, the number of projects times their number of changes takes a lot of time. And that's why I am using a big machine. Currently, I am making use of ~10 cores and a lot of RAM!
Nour Assy Postdoc researcher at the AIS group of TU/e
I work on process variability mining. I develop techniques to mine process models from large repositories of event logs. The mined models are often very large and complex. To evaluate the quality of my models I need to compute their precision and recall. I am using computationally and memory expensive algorithms, that's why I am using the BSR machine at TU/e.