The purpose of this report is investigate the usage of algorithms aimed to optimize big data cluster configurations. Such a configuration might have many properties, and three different algorithms is presented with different approaches aimed to find an optimal configuration based on a set of rules. Presented is a suggested model and an approach based on simulations. The results from these simulations indicate that using such an approach can be viable in finding a point in a multidimensional space with dimensions representing a vector of a configuration property such as the number of CPUs or the amount of memory per node. There is more work to be done in this field of study. A model has been suggested, however the implementation and installation on a real life big data cluster is outside the scope of this report and would certainly be useful.