“Our project had unforeseen challenges. One of which was the fact that we needed to extend the capabilities of the simulator for us to be able to run our revolutions,” said Stewart. “Another challenge was that we needed to run our computationally expensive, lengthier simulations on a high-performance computing cluster.
“However, our project code was in C++, meaning it needs to be compiled where one intends to run it. In short, our project had multiple dependencies and required multiple external libraries, when all added together it means that the process itself would be an exceptionally time-consuming task. We realised too late that we did not have the admin privileges required to install what was needed on the clusters, meaning that we wouldn’t have been able to complete the project within the time available,” she added.