This paper describes the implementation of high performance process simulator clusters based on modern graphical processing unit architectures. Graphics rendering is basically just an intensive session of complex mathematical calculations, reading and writing memory buffers and computing bitmap matrixes. This is extremely similar to what system simulation software actually does: some mathematics - less complex than the usual graphics rendering in a typical 3D game, updating some algorithm state variables held in memory, and computing an output for the system. The first part of this article will focus on comparing hardware architectures between the central processing unit and graphics processing unit, and the second part will focus on implementing a PID controller on both. A case study will be presented in the last part, presenting a highly efficient mobile application which will employ the on-board graphics processing unit to provide system simulations while also saving energy.