In the field of cell biology and biochemistry, it is now widely accepted that spatiotemporal dynamics play vital roles in cellular systems such as signal transduction and cell cycle control. For an efficient engineering of biochemistry with such problems, we have been developing a modeling system based on Monte Carlo simulation methods. In this system, protein molecules are described as particles undergoing continuously the Brownian motion in a 2D/3D space. Such description allows one to analyze the spatiotemporal problems in a self-consistent manner for which most of the conventional simulation studies did not pay much attention. In a perspective of a full modeling of the whole cell pathways, we introduce our development of a special purpose hardware for the modeling system. This hardware is estimated to enable more than 100-fold speedup from the latest high-performance personal computers