Numerical simulations of combustion processes are usually carried out with many simplifications in order to keep the computing effort low. However, if flames are to be simulated with a high degree of detail, many physical processes must be taken into account. Such detailed simulations are only possible on supercomputers. The paper Optimizing Load Balancing of Reacting Flow Solvers in OpenFOAM for High Performance Computing presents a method that accelerates the parallel calculation of combustion processes on supercomputers. When calculating chemical reaction rates, it is often the case that some processes require more computing effort than others. The basic idea, therefore, is to form pairs of processes that share their work of calculating chemical reaction rates. This is made possible by the fact that the calculation of chemical reaction rates can be more freely distributed among the processes than the calculation of many other simulation variables. The methodology was implemented with the open source CFD software OpenFOAM, which itself does not provide load balancing for the calculation of chemical reaction rates. The paper of the SimLab Energy was awarded the Best Paper Award at the 6th OpenFOAM Conference 2018 in Hamburg.
Thorsten Zirwes