The particle simulation library AutoPas implements many algorithms with vastly different performance characteristics to solve molecular dynamics simulations. During the simulation, it uses black-box optimization techniques to automatically select the fastest algorithm for the current simulation state. While they are able to find good algorithms eventually, they often try highly unsuitable algorithms at the start. As some algorithms perform orders of magnitude worse than the optimum for a given simulation state, this has a great negative impact on the time to solution.

In this project, we gather domain-specific knowledge about the performance characteristics of the algorithms through theoretical modeling, profiling, and benchmarking. We make the results available through a new white-box optimization strategy that is able to apply any domain-specific knowledge during optimization. Furthermore, we give insights and recommendations what can be done to potentially improve the performance further.